Conference Agenda

Semantic 3D indoor mapping often depends on supervised learning and large annotated datasets, limiting scalability across diverse environments. This work introduces a category-specific prompt strategy for semantic 3D mapping using RGB-D cameras, integrating RGB-D SLAM with the Segment Anything Model 2 (SAM2) to enable annotation-efficient reconstruction. Keyframes and trajectories extracted from SLAM provide spatial references, while SAM2 performs zero-shot segmentation guided by a Category-Specific Prompt Strategy (CPSS), which segments structural and functional elements (e.g., floors, doors, staircases) by category to reduce prompt interference and manual effort. The segmented keyframes are then fused with depth and pose data to produce instance-level semantic point clouds. Experiments on custom RGB-D sequences and selected ScanNet scenes demonstrate centimeter-level geometric accuracy and strong semantic consistency, with mIoU values up to 0.89 on the custom dataset and 0.98 on ScanNet. The resulting semantic point clouds are clean, structured, and require minimal post-processing, showing that the proposed strategy provides an efficient and scalable solution for semantic 3D indoor mapping without retraining or environment-specific supervision.

Mobile robots are widely used in unmanned surveying, warehouse logistics, and emergency response. However, achieving safe, reliable, and efficient autonomous navigation in unknown environments remains challenging, where accurate environment representation and feasible trajectory planning are crucial. This paper presents an autonomous navigation method integrating lightweight LiDAR mapping with real-time local planning for ground robots. At the perception level, an incremental single-frame point cloud update is used to accumulate and project locally traversable space, producing a lightweight obstacle map that preserves geometric accuracy while reducing planning computation. At the planning level, A* is employed to generate reference control points, and uniform B-spline curves are used to optimize the trajectory while enforcing kinematic feasibility and smoothness. At the control level, nonlinear model predictive control (NMPC) ensures accurate trajectory tracking by producing control commands that satisfy velocity and acceleration constraints. The framework also supports low-cost evaluation in simulation. Experiments in simulated forests, simulated indoor corridors, and real-world gardens and hallways show average navigation speeds of 2.24 m/s, 0.76 m/s, 0.43 m/s, and 0.38 m/s, respectively. Results demonstrate that the proposed method generates smooth, feasible, and safe trajectories and completes autonomous navigation and mapping tasks across diverse environments.

Place recognition is a crucial task for both robots and autonomous vehicles, facilitating positioning and loop closure within pre-built global maps. Although single-modal sensor-based methods have shown satisfactory performance, cross-modal place recognition—retrieving low-cost camera images from global point cloud databases—remains a significant challenge. In this paper, we propose a contrastive learning-based lightweight cross-modal place recognition framework (CMCL-PR) to retrieve a single image from a global offline point cloud map. We introduce a perspective projection based field-of-view(FoV) transformation module that converts point clouds into a modality analogous to images; Then, we design a dual branch intra-modal encoder structure based on shared Transformer, which extracts and aligns image and point cloud features separately, effectively unifying the feature distribution between modals; Besides, a cross-attention mechanism module guided by inter-modal consistency was constructed, which utilizes the contribution of scene context information within different modalities to generate the discriminating cross-modal descriptors. Finally, during the contrastive learning, cross-modal feature was enhanced, and a multi loss function was constructed, including cross-modal contrastive learning loss, intra-modal consistency loss, and matching supervision loss. We assess the effectiveness and generalizability of our method using three publicly available datasets: KITTI, KITTI-360, and Oxford RobotCar. The project page and code will be released at https://github.com/qp-li/CMCL-PR.

The presence of refraction-induced systematic errors has always been a cause of concern in the field of underwater photogrammetry.

This work extends previous studies from the authors with new simulations specifically aimed at practical applications underwater using popular sensor devices and configurations, such as GoPro action cameras fitted with standard flat port housing that are very common among marine ecologists and archaeologists.

We aim at investigating whether approaches used by regular photogrammetry above water can be applied underwater without a significant accuracy loss for the application of interest.

Due to the complexity of collecting ground truth data, simulations are used. We utilize the POSER framework (https://github.com/GEOSS-UNISS/POSER) developed within the 2024 ISPRS Education and Capacity Building Initiatives (ECBI).

We investigate the benefits and cons of the refractive vs the implicit modelling approaches with respect to estimability of camera calibration (refractive) parameters, need for pre-calibration setups with approaches from literature, availability of ground control points, and assessing the accuracy of both approaches against ground truth simulated data.

The accuracy is reported as discrepancies in the reconstructed 3D models, exterior orientation and camera calibration parameters.

Reconstructing the water surface in refractive domains such as rivers and lakes is challenging, since light bending at the air-water interface alters the apparent geometry and breaks the straight-ray assumption of conventional image-based 3D reconstruction. Accurate water surface models are therefore a key prerequisite for many refraction-aware applications. This contribution investigates the potential of three passive image-based methods, Structure from Motion (SfM), Multi-View Stereo (MVS), and Monocular Depth Estimation (MDE), to derive a geometrically consistent water surface model from UAV imagery of the Pielach River study site in Austria. The dataset represents a demanding scenario with clear, fast-flowing water and low texture, which causes strong refraction and poor feature stability. Quantitative comparisons against LiDAR-derived reference surfaces show that SfM yields sparse and inconsistent points, MVS reconstructs the riverbed instead of the water surface, and MDE exhibits scale and offset inconsistencies despite explicit calibration using SfM reprojections. Completeness remains below 45 % for all methods with mean vertical deviations in the decimetre-to-metre range. The results indicate that current image-based approaches are insufficient for reliable water-surface reconstruction in such settings, reinforcing the need for an explicitly derived surface model as input to refraction-aware modeling, for example in bathymetric reconstruction and future refractive neural rendering methods, rather than relying on implicitly learned water surfaces.

This article studies the complementary usage of RTI and SfM-MVS visual inspection (visual testing) of welds under water. Two compact low-cost lighting domes of different designs were developed and deployed with a monochromatic camera at close range.

The lighting domes generate homogenous and direct illumination, respectively, tailoring the requirements of SfM and RTI. The camera is housed in a cylindrical tube, equipped with a dome port interface.

The 3D reconstruction in combination with RTI models could augment existing testing strategies and provide digital, gapless documentation. Experiments were conducted in laboratory in air, clear and turbid water questioning were the capabilities and limits are for given setup with respect to visual testing of welds.

Under the correct lightning, in air the techniques perform on a high accuracy level and are well suited for inspecting welds digital and interactively. Underwater the results differ in dependence of the degree of turbidity and prove to be sensitive for configurational parameters leaving space for improvements of acquisition and processing workflows. However, even in turbid water the 3D reconstructions and RTI models could be calculated enabling novel possibilities for weld inspection.

We present the first quantitative evidence of systematic daily dead tree branch movement under field conditions, which has been previously only observed qualitatively. Using a boreal forest Permanent Laser Scanning (PLS) dataset, we investigated whether such movement can be detected and characterized in 3D laser scans. We link branch anatomy and geometry with environmental drivers, leading to the hypothesis that branch bending is proportional to wood moisture content. Hourly point clouds collected over 3.5 days from 17 dead branches (16 coniferous), attached mainly to living trees, were analyzed under calm weather conditions. We developed a novel workflow that tracks movement via non-rigid registration, calculates the angle between each branch and the vertical, and quantifies branch movement uncertainties across epochs. After accounting for time lags, these movements were related to the modeled wood moisture content using a linear mixed model. Clear and consistent daily movement was detected in all branches, with a mean amplitude of 21 cm and an average delay of 3.5 hours relative to moisture content changes. All branches moved downward during the day and upward at night, except one deciduous branch displaying the opposite pattern, consistent with our theoretical framework. The linear mixed model revealed a significant positive linear relationship between branch movement and wood moisture content. Our findings confirm that daily dead tree branch movement is primarily hygroscopic and demonstrate its effective detection using operational PLS. These insights open new possibilities for monitoring tree vitality using hypertemporal 3D sensing.

Tree growth determines how much CO2 is sequestered from the atmosphere and temporarily stored in woody biomass. At the same time tree growth is affected by increasing temperatures, more frequent drought periods, late frosts and other extreme events associated with climate change. While continuous measurements of radial (secondary) tree growth using dendrometers are well established, monitoring of shoot elongation (primary growth) has largely been neglected because suitable measurement techniques are lacking. As a result, the effects of climate change on primary tree growth remain insufficiently understood. This work aims at reconstructing native deciduous trees in 3D as a basis for measuring and monitoring shoot elongation over entire tree canopies. Here we explored the use of low-cost UAV photogrammetry and of a multi-camera CraneCam system under real-world conditions. Data were collected in two study areas over an entire growing season. We present sensor evaluations, photogrammetric data acquisition and processing strategies. A special focus is placed on the analysis of the resulting photogrammetric 3D point clouds in terms of accuracy, resolution and completeness. Results demonstrate 3D point accuracies of 5-6 mm for entire trees using consumer-grade UAVs weighing less than 250 g and a 3D reconstruction completeness between 92% and 98% depending on the UAV type. The paper introduces a novel 3D-printed ground-truth branch to evaluate the capability to reconstructing fine-detail structures such as thin tree shoots. Finally, we discuss operational challenges and initial experiments towards a skeletonization of entire trees based on photogrammetric point clouds.

Urban trees provide critical cooling benefits and regulate microclimates in cities, yet their three-dimensional thermal behaviour remains difficult to visualize and communicate to stakeholders. Traditional approaches rely on 2D thermal imagery analyzed on desktop screens, lacking spatial context and the capability for temporal monitoring across phenological stages. This work presents a pipeline combining Gaussian Splatting with augmented reality to enable immersive, on-site visualization of urban tree thermal patterns.

We captured thermal infrared (TIR) images of silver linden trees (Tilia tomentosa) at the University of Strasbourg using a FLIR T560 thermal camera. The TIR images were preprocessed and then processed using MILo (Mesh-In-the-Loop Gaussian Splatting), generating 3D thermal meshes with approximately 3 million vertices. The reconstruction is validated against reference thermal point clouds acquired with a terrestrial laser scanner equipped with an integrated thermal camera, assessing both geometric completeness and thermal attribute fidelity.

The thermal meshes are deployed in a mobile augmented reality application, allowing users to visualize temperature distributions directly overlaid on physical trees in the field. This work demonstrates the first application of 3D Gaussian Splatting to thermal vegetation modelling, providing an engaging educational tool to communicate urban trees' cooling role while offering researchers a platform for detailed thermal analysis and forest health monitoring.

To robustly support glacier calving monitoring at high temporal resolution and enable future AI-based calving forecasts, this study presents an optimized Multi-Epoch Multi-Imagery (MEMI) strategy for automated 4D point cloud model generation. To date, the dataset comprises over 160,000 images acquired since December 2024 by an autonomous multi-camera system operating at 30 min intervals at Glacier Perito Moreno (GPM), Argentina. Despite high scene variability and harsh environmental conditions, the proposed MEMI workflow effectively addresses constraints imposed by continuous glacier motion and image degradation. The enhanced strategy aims to generate precise dense clouds with high alignment accuracy and computational efficiency, forming the basis for subsequent analysis of glacier front evolution. To achieve this, various parameter configurations are evaluated, including AI-based image masking and adaptive, optimized alignment-adjustment settings. Results from a representative eight-day subset show that variations in the tie point computation strategy lead to measurable differences in alignment-adjustment efficiency, with the best configuration being about 11 % faster than the least efficient one. By contrast, adaptive alignment-adjustment consistently improves alignment accuracy. Moreover, masking enhances both image quality checking and reconstruction quality, and, albeit modestly, improves pre-failure deformation analysis. Furthermore, daily seasonal responses to alignment are observed, as accuracy varies with solar illumination relative to the camera positions. Applying the optimal configuration to 260 MEMI projects in under 42 h produced 518 high-precision dense clouds and detected calving retreat magnitudes of up to 17.5 m, demonstrating the robustness and scalability of the proposed MEMI strategy for high-temporal-resolution 4D point cloud generation.

Gorgan Bay in the southeast of the Caspian Sea faces significant issues with water volume reduction and water quality deterioration. The turbidity levels of this water body have increased recently owing to the ongoing decline in the Caspian Sea level and the increase in human activity. In this study, to monitor water quality of the bay, various machine learning models were used to retrieve turbidity levels from Sentinel-2 satellite imagery. In situ turbidity measurements acquired throughout the bay were correlated with Sentinel-2 reflectance data. A statistical evaluation was conducted to ascertain the prospective band combinations for estimating turbidity. Four regression methods, including Multiple Linear Regression (MLR), Random Forest (RF), Classification and Regression Trees (CART), and Gradient Tree Boost (GTB), were implemented to estimate turbidity levels using six different input scenarios. These models were tested on unseen test data, and it was found that the CART model with RMSE = 7.89 FTU, R² = 0.93, and Nash-Sutcliffe efficiency (NSE) = 0.74 exhibited superior performance. The generated turbidity maps across the bay showed sediment plumes next to southeastern river mouths, indicating increased turbidity levels in these areas compared to the rest of the bay, revealing intra-bay variability due to tidal and discharge dynamics. The applied methodology demonstrated superior performance compared to conventional empirical models in turbid coastal environments. The results indicated that the machine learning approaches coupled with satellite data provides water resource managers with a cost-effective and real-time tool for coastal water quality monitoring.

The formation of pit lakes in decommissioned open-pit mines has raised concerns regarding long-term water quality. Turbidity, a key indicator of suspended particulate matter, influences water clarity and aquatic ecological processes. This study estimates surface turbidity in the Águas Claras Mine (MAC) pit lake in Nova Lima, Brazil, using satellite imagery and in situ data to generate a continuous time series and assess compliance with thresholds established by current Brazilian environmental legislation (CONAMA Resolution No. 357/2005). Landsat 5 and 8 imagery were used to derive a spectral turbidity index. Based on the temporal overlap between satellite and field data, a linear regression model (R² = 0.77) was developed and applied to extend the turbidity time series. The results indicate that turbidity values remained below the legal limits for Class 1 freshwater. Higher turbidity levels were observed during the initial filling phase, associated with exposed slopes, as well as episodic increases during the rainy season due to sediment runoff. Over time, progressive revegetation and minimal anthropogenic disturbance contributed to the stabilization of water quality conditions. The integration of in situ measurements and remote sensing proved to be an effective approach for monitoring water quality in post-mining environments, supporting both environmental liability assessment and closure management.

Harmful algal blooms caused by Margalefidinium polykrikoides have intensified in Korean coastal waters, yet existing bio-optical models are not able to reproduce the species-specific spectral features required for quantitative bloom assessment. This study develops a dedicated semi-analytical bio-optical model by integrating multi-year field measurements collected from six campaigns between 2013 and 2022, including hyperspectral above-water radiometry, laboratory absorption spectra, and chlorophyll-a (Chla) observations.

The model formulation follows a standard absorption–backscattering reflectance framework, in which total absorption is decomposed into water, phytoplankton, NAP, and CDOM components, while phytoplankton backscattering is parameterized using two optimized species-dependent parameters. An iterative inversion procedure identifies the optimal backscattering structure by minimizing the spectral mismatch between modeled and measured hyperspectral Rrs. In addition, an empirical red-edge term is introduced to capture the distinct fluorescence-associated peak near ~700 nm that characterizes high-biomass M. polykrikoides waters.

The resulting model accurately reconstructs observed Rrs across low to high Chla conditions, reproducing key features such as strong blue absorption, the secondary blue rebound, and the pronounced red-edge peak. Comparisons with GIOP and Karenia-based models show substantially improved performance, particularly under extreme bloom conditions.

This work provides the first validated species-specific bio-optical parameterization for M. polykrikoides and offers a practical pathway for satellite-based HAB monitoring using upcoming hyperspectral missions such as PACE and GLIMR. The framework is extendable to additional HAB species and supports future development of physics-based, species-resolved coastal water-quality retrievals.

Coastal zones are highly vulnerable to multiple hazards, including tsunami, shoreline erosion, coral reef degradation, and escalating impacts of sea-level rise. These concerns illustrate the urgent need for accurate and high-resolution geospatial data in coastal areas are required to support coastal risk assessment and management. A seamless and accurate coastal digital elevation model (DEM) is a foundational dataset to support these needs.

However, the development of a seamless land-sea elevation surface remains challenging. The intertidal zone often forms a critical data gap between land DEMs and bathymetry grids. To address these limitations, the use of multi-sensor geospatial data has grown considerably in coastal science and hydrography, such as structure-from-motion (SfM) photogrammetry and satellite-derived bathymetry (SDB). Assimilating SfM and SDB offers a viable pathway for constructing seamless coastal DEM.

Therefore, understanding the quality of SfM and SDB data before integration is a critical step. This study addresses this gap by evaluating the vertical accuracy, effective spatial resolution, and internal consistency of SfM-derived coastal topography and SDB-derived shallow-water bathymetry in a challenging coastal environment commonly found in Indonesian waters, i.e., coral reef islands. In this study, the area of interest is Karimunjawa and Kemujan Islands, located in the Java Sea, Indonesia, approximately 80-90 km from the mainland.

Based on our preliminary results, SfM provides high spatial detail depths but introduces short wavelength oscillation while SDB shows smoother depth gradients. In addition, several depths derived from SfM and SDB indicate different vertical reference levels.

This presentation introduces a digital twin and immersive VR system that reconstructs the “Hidden Foot” of Borobudur Temple in Central Java, Indonesia—an encased architectural layer containing 160 narrative bas-relief panels that cannot be directly observed today. The system integrates multi-source 3D point cloud data acquired via UAV and ground-level imaging with deep-learning–based 3D reconstructions of reliefs from historical monocular photographs and semantic segmentation results from previous work.

A custom passage model simulates the narrow gap between the original relief-bearing wall and the surrounding protective masonry, allowing users wearing a head-mounted display to virtually enter and walk through this otherwise inaccessible space. Within the VR environment, users can switch between photorealistic relief rendering and color-coded semantic overlays to support both immersive appreciation and analytical interpretation. A torch-based Level of Detail (LoD) strategy, driven by dynamic illumination, maintains high visual fidelity near the user while reducing rendering load for distant geometry, ensuring stable frame rates suitable for comfortable VR exploration.

A small user study indicates high ratings for visual realism, immersion, and the educational clarity of semantic overlays and gaze-triggered text annotations, highlighting the potential of this approach for research, documentation, and public engagement with hidden cultural heritage.

Three-dimensional (3D) scanning is widely used to preserve cultural heritage as large-scale point clouds. While these datasets contain rich geometric information, transparent visualization of such massive point clouds often suffers from visual clutter and reduced clarity, particularly when both external and internal structures are involved. Previous work resolved the problem of normal orientations, laying the foundation for robust shading in transparent visualization. Building on this foundation, this paper introduces a novel method of adaptive opacity control for region highlighting, which interprets shading as a distribution of opacity. By adjusting the lighting direction, effective opacity can be locally controlled without modifying the original point cloud data. This mechanism enables selective highlighting of user-specified regions, enhances the visibility of complex structures, while also allowing interactive dynamic shading by continuously changing the lighting direction. The effectiveness of the proposed method is demonstrated using culturally significant heritage point clouds, including UNESCO World Heritage sites, where intricate internal structures can be more clearly analyzed. Beyond cultural heritage, the proposed method is also applicable to modern architectural and other large-scale 3D scanned objects with similarly complex forms.

The concept of Digital Twin has attracted growing interest within research communities, including heritage conservation, in recent years. It combines detailed geometric documentation, real-time monitoring, and semantic information to create dynamic digital replicas of historic buildings. This paper presents the results of a scoping review of 204 peer-reviewed studies published between 2020 and 2025. The aim is to identify the main technologies, devices, and methods used to develop a Digital Twin for heritage buildings. The review reveals that terrestrial laser scanning (TLS), UAV photogrammetry, BIM, and IoT sensor networks form the core technological base. It also highlights the growing use of artificial intelligence for automated defect detection, predictive maintenance, and semantic processing. Based on the reviewed literature, the paper introduces a six-stage workflow for building a heritage Digital Twin, covering baseline documentation, static reality capture, semantic modelling, sensor integration, data fusion, and operational use. The findings show a clear shift from static 3D documentation toward dynamic, data-rich systems that support continuous monitoring and more informed decision-making. However, the review also identifies major challenges, including limited interoperability, complex data integration, incomplete AI validation, and long-term digital preservation issues. Overall, the study outlines the current state of Digital Twin technologies in architectural heritage and identifies key areas that require further research to support reliable and sustainable applications.

In the current context of energy and climate transition, the preventive conservation of historic buildings is particularly important due to their impact on architecture and works of art. Establishing the correlation between environmental variables and the condition of artworks in situ requires comprehensive and individualized monitoring, allowing for an understanding of cause-and-effect mechanisms. To address this challenge, the EPICO method provides a systematic framework for assessing deterioration risks in palace-museums through multi-scale monitoring and correlation between environmental parameters and object condition. The aim of the proposed topic is to enhance this decision-making tool with the creation of digital twins. These digital twins being fed with three-dimensional hyperspectral and LiDAR mapping of spaces and objects.

Urban digital twins are virtual city replicas that can greatly support urban planning by simulating infrastructure and mobility scenarios. However, keeping a digital twin up-to-date with fine-grained, real-world urban conditions is challenging. This paper proposes a novel system that leverages multi-modal AI models to bridge the gap between physical urban data collection and a 3D city digital twin. In our approach, ordinary smartphones carried in vehicles act as mobile sensors, continuously capturing multi-modal data (road images, GPS coordinates, and speed). Advanced vision-language models then analyze the data to automatically extract information from the traffic infrastructure and detect road anomalies. The extracted information such as the locations of traffic signs, traffic signals, road surface cracks, and potential blind spots at intersections is geo-tagged and streamed into language-vision models to interpret data and stream human readable insights into the digital twin model. The case study is the digital twin of the city of Toronto. By aggregating data from many drivers and analyzing it (in post-processing for high accuracy), the digital twin evolves into a living model of the urban environment. This enriched and dynamic twin provides urban planners with up-to-date insights on traffic signage, road conditions, and other relevant road infrastructure elements, enabling proactive maintenance and informed decision-making for city planning.

Methods that synthesize indoor 3D scenes from text prompts have wide-ranging applications in film production, interior design, video games, virtual reality, and synthetic data generation for training embodied agents. Existing approaches typically either train generative models from scratch or leverage vision-language models (VLMs). While VLMs achieve strong performance, particularly for complex or open-ended prompts, smaller task-specific models remain necessary for deployment on resource-constrained devices such as extended reality (XR) glasses or mobile phones. However, many generative approaches that train from scratch overlook the inherent graph structure of indoor scenes, which can limit scene coherence and realism. Conversely, methods that incorporate scene graphs either demand a user-provided semantic graph, which is generally inconvenient and restrictive, or rely on ground-truth relationship annotations, limiting their capacity to capture more varied object interactions. To address these challenges, we introduce GeoSceneGraph, a method that synthesizes 3D scenes from text prompts by leveraging the graph structure and geometric symmetries of 3D scenes, without relying on predefined relationship classes. Despite not using ground-truth relationships, GeoSceneGraph achieves performance comparable to methods that do. Our model is built on Equivariant Graph Neural Networks (EGNNs), but existing EGNN approaches are typically limited to low-dimensional conditioning and are not designed to handle complex modalities such as text. We propose a simple and effective strategy for conditioning EGNNs on text features, and we validate our design through ablation studies.

The Low-Altitude Economy (LAE), driven by the widespread deployment of UAVs and eVTOL aircraft, demands a high-fidelity Digital Twin that extends far beyond static geographic representation. This study presents a critical review of 39 peer-reviewed papers to propose a three-layer mapping framework — Geospatial Infrastructure Layer, Environmental Sensing Layer, and Interaction Layer — and evaluates the Technology Readiness Level (TRL) of each sub-domain. The Geospatial Infrastructure Layer encompasses terrain models, ground facilities, airspace structures, and semantic navigation landmarks. The Environmental Sensing Layer covers electromagnetic modeling, target sensing and countermeasure, and micro-meteorological mapping. The Interaction Layer addresses network trust, data security, swarm coordination, and platform reliability. Our TRL assessment reveals that Environmental Sensing is the most mature layer (mean TRL 4.4, 8 field-validated papers), while cross-layer integration remains the weakest link (mean TRL 3.5, zero field-validated demonstrations). We identify standardization of low-altitude spatial data products, AI-enabled predictive mapping, crowdsourced Digital Twin updating, and closed-loop cross-layer integration as the four priority research directions.

Indoor pedestrian localization on smartphones requires a lightweight yet robust fusion framework capable of handling unstable wireless signals and drift-prone inertial motion. In this work, we propose an efficient positioning system that integrates weighted Wi-Fi fingerprinting with Pedestrian Dead Reckoning (PDR) through a Multi-Step Extended Kalman Filter (EKF). Unlike conventional single-step filtering, the EKF employed here is formulated from the perspective of factor-graph–based probabilistic estimation, where Wi-Fi observations and PDR increments naturally act as complementary measurement and motion factors. Building on this unified view, the proposed Multi-Step EKF retains several consecutive states within its estimation window, effectively approximating short-horizon smoothing while maintaining the computational footprint required for real-time execution on consumer smartphones.

To enhance observation stability, an inverse-distance weighted fingerprinting module mitigates RSSI fluctuations and gracefully handles missing values. Meanwhile, PDR inputs are refined through diagnostic analysis and incorporated as motion constraints within the fusion process. Global optimization of noise parameters is performed via dual annealing, further improving the reliability of state updates.

Experiments conducted on an open indoor dataset demonstrate that the proposed method achieves a reduction of approximately 31% in mean localization error compared with a standard single-step EKF baseline. The results confirm that enforcing short-term temporal consistency through a multi-step state representation significantly suppresses drift accumulation and enhances robustness under dynamic indoor environments. Overall, the proposed framework offers a theoretically grounded, computationally efficient, and practically deployable solution for smartphone-based indoor positioning.

This project, a collaboration between the Landesamt für Geobasisinformation Sachsen (GeoSN) and M.O.S.S. Computer Grafik Systeme GmbH, pioneers the development of a Digital Twin Forest prototype for Saxony. The initiative leverages high-resolution aerial orthophotos (DOP) and advanced AI methods to generate detailed, current forest information. The core methodology centers on the “DeepTrees” workflow, a convolutional neural network (CNN)–based approach developed by the Helmholtz Center for Environmental Research (UFZ). This workflow processes DOP imagery at 10–20 cm resolution to segment individual tree crowns and extract key forest indicators, including crown area, crown radius, and tree density.

The process unfolds in three main stages: (1) preprocessing and model adaptation using transfer learning, (2) inference and postprocessing for accurate tree segmentation, and (3) integration into GeoSN’s data infrastructure. This integration utilizes OGC-compliant services and moGI-based data management, enabling automated processing, configuration, and visualization.

Results from the prototype confirm the feasibility of precise, large-scale tree crown segmentation from aerial imagery. The system also demonstrates the potential to derive temporal and structural forest information from recurring DOP datasets. These outputs can be directly incorporated into operational geospatial systems, supporting climate adaptation, forest management, and policy-making.

In conclusion, the Project showcases how explainable, interoperable AI workflows can strengthen national geodata infrastructures and serve as a model for future federated, AI-driven digital forest twins across Germany.

This study presents a scalable framework for mapping aboveground biomass and moisture content in peatlands using intensive field sampling, multi-sensor satellite imagery, and advanced machine learning. Field data collected from diverse bog and fen sites in Western Newfoundland are integrated with Sentinel-1/-2 synthetic aperture radar and optical data, complemented by 3 m PlanetScope imagery for site-level detail. Ensemble learning models, particularly XGBoost, yield high biomass mapping accuracy, with regional maps revealing major biogeographical gradients and fine-scale site mosaics. Feature importance analysis highlights the role of red-edge and SAR bands in prediction. The results demonstrate that free satellite archives and machine learning can overcome limitations of costly airborne campaigns, supporting operational carbon monitoring and ecological management in northern peatlands. This approach establishes a foundation for wide-area wetland monitoring and future expansion using emerging remote sensing technologies.

Surface soil moisture (SM) plays a significant role in environmental and hydrological processes, particularly runoff and evapotranspiration. Within forest ecosystems, changes in SM can lead to significant ecological impacts, including paludification and greater susceptibility to forest fires. Microwave remote sensing facilitates large-scale monitoring of SM. Moreover, machine learning (ML) have demonstrated strong potential for capturing the nonlinear relationships between SM and satellite data. In general, supervised ML techniques achieve higher success rates when trained on larger ground measurements. However, obtaining extensive ground measurements of SM over vast areas such as forests is challenging, expensive, and time-consuming. To address this limitation, this study proposes a self-supervised method based on pre-task learning to estimate SM over forested areas using multisensor data. The core idea of the self-supervised approach is to leverage the knowledge gained during pre-task learning from multisensor data and transfer it to the SM estimation task, thereby improving the model’s generalization ability to SM estimation. The self-supervised learning method achieved an overall coefficient of determination (r²) of 0.74 and an RMSE of 0.04 m³/m³ on the testing dataset By focusing on each forest site, the model obtained r² = 0.75 with RMSE = 0.04 m³/m³ at Millbrook, r² = 0.63 with RMSE = 0.04 m³/m³ at Massachusetts, and r² = 0.74 with RMSE = 0.03 m³/m³ at Saskatchewan. The results highlight the potential of multisensor data for SM estimation in forested areas. Our method, which utilizes self-training on the input data, reduces dependence on ground SM measurements and enhances generalization capability.

Land cover data need to be updated regularly. Typically, remote sensing images (RSI) play a central role in this process. A first step is RSI semantic segmentation. Today, this task is mainly solved by deep learning. Especially vision foundation models (VFM) have gained increasing importance in this context. Having been trained on large datasets, VFM for segmentation can yield good results on data from various domains without further training. We present a new method for using the VFM Segment Anything Model 2 (SAM 2) for multi-class semantic segmentation of Sentinel-2 images that does not require training data. Our method is based on a prompt engineering approach, using SAM 2 in its pre-trained form. The different prompt types are generated on the basis of existing topographic data. We also propose a post-processing step for merging the output of SAM 2 to obtain a multi-class label image. The results of our experiments show that our method achieves an overall accuracy (OA) of up to 93% at pixel-level using mask prompts. Experiments with other Sentinel-2 3-channel composite images do not show significantly different results compared to R-G-B images. Incorporating data from different time steps, intended to be used for map updating, shows good results. But the small amount of changed areas indicate limitations. In general, the proposed method is suitable for further

research into semantic segmentation tasks with little or no training data, as well as for the process of updating databases.

This paper presents a radar-centric multi-sensor fusion framework, RLIO, designed for robust indoor SLAM in perceptually challenging environments such as underground garages and smoke-filled areas. Unlike conventional LiDAR-based methods that degrade under poor visibility, RLIO tightly integrates 4D imaging radar, 3D LiDAR, and an IMU within an iterated extended Kalman filter. The system introduces three key modules: a motion-prior-driven radar velocimetry algorithm for stable velocity estimation, a velocity-prior-enhanced scan-to-map registration for drift reduction in degenerate geometries, and an adaptive fusion strategy that dynamically adjusts sensor weights based on real-time degradation detection. Experimental results from both handheld and UGV platforms demonstrate that RLIO achieves accurate localization and high-quality mapping even when LiDAR performance deteriorates due to smoke or repetitive structures, highlighting its potential for reliable all-weather autonomous navigation and mapping in complex indoor and outdoor environments.

4D radars have attracted increasing interest for robotic perception because they remain effective in adverse conditions such as darkness, dust, smoke, rain, and fog. Compared with conventional automotive radars that mainly provide planar coordinates and relative Doppler velocity, modern 4D radars also sense elevation, which makes them more suitable for geometric odometry and mapping.

In this paper, we propose RAMBA, an offline 4D radar mapping framework based on bundle adjustment. Starting from initial poses and radar frames produced by a radar--inertial odometry front-end, we refine the radar frame states to improve global mapping consistency, measured by covariance-weighted point-to-point distances. In essence, our method extends pairwise generalized iterative closest point (GICP) to the multi-frame setting. Candidate correspondences are formed within voxels of a voxel grid built from all selected frames, and each residual is weighted by the sum of the two point covariances. The geometric constraints are jointly optimized with IMU preintegration and radar ego-velocity constraints.

To reduce false associations caused by drift and revisits, RAMBA enforces temporal consistency when forming correspondences and explicitly allows constraints around loop closures. We evaluate the method on the ColoRadar and SNAIL Radar datasets. The proposed refinement consistently improves map quality and usually improves trajectory accuracy over the initial radar--inertial odometry and pose graph optimization. To the best of our knowledge, this is the first geometric offline bundle-adjustment framework for consistent 4D radar mapping.

Four-dimensional (4D) imaging radar offers robustness in adverse weather and lighting, but its point clouds remain sparse, noisy, and affected by ghost reflections, making geometric scan matching unstable. This work integrates two existing deep correspondence models—Radar Transformer and RPM-Net—into a radar–inertial odometry pipeline without retraining. Both networks run asynchronously in dedicated ROS nodes: radar and submap point clouds are cropped, transformed, and sent to the matchers, which return either hard or soft correspondences for the first IEKF iteration of each frame. When neural outputs are delayed, the system automatically falls back to geometric matching.

Returned matches are fused with a voxelized IEKF backend that computes Mahalanobis-weighted residuals. RPM-Net further supplies soft targets and confidence weights, enhancing point-to-point constraints. Experiments on ColoRadar indoor and outdoor sequences show that learning-based correspondences can reduce drift in weakly structured scenes while maintaining robustness when geometry is reliable.

This paper presents MSMR SLAM, a multi sensor and multi robot SLAM framework that integrates LiDAR, 4D radar, and IMU to achieve robust localization and consistent mapping in large scale and degraded environments. The front end includes radar inertial odometry and LiDAR inertial odometry, and evaluates the reliability of LiDAR observations using point cloud sparsity and effective range. An adaptive fusion module combines the two odometry estimates to maintain stable state estimation, while radar assisted dynamic point removal improves the reliability of geometric constraints. The back end constructs a unified factor graph that incorporates multi sensor odometry constraints, loop closure factors, and inter robot association factors. A LiDAR centered and radar assisted matching strategy enhances cross robot data association, and radar based loop closures improve global consistency when LiDAR measurements degrade. The system maintains a dense LiDAR map together with a complementary radar map, enabling hybrid mapping that remains reliable in perceptually challenging regions. Experiments on campus datasets, including smoke filled scenarios, demonstrate that MSMR SLAM achieves high precision multi robot localization and globally consistent mapping. Compared with single robot baselines, the proposed framework provides improved accuracy and robustness, and the integration of LiDAR and radar yields more complete and stable map reconstruction in complex environments.

1. We introduce a novel framework for 4D radar-camera extrinsic calibration that utilizes commonly available dynamic objects as natural correspondences.

2. We develop a heuristic-guided strategy to reliably associate radar points with image detections and estimate the extrinsic parameters without deep learning.

General-purpose spatial representations are essential for building transferable geospatial foundation models (GFMs). Among them, the AlphaEarth Foundation (AE) represents a major step toward a global, unified representation of the Earth's surface, learning 10-meter embeddings from multi-source Earth Observation (EO) data that capture rich physical and environmental patterns across diverse landscapes. However, such EO-driven representations remain limited in capturing the functional and socioeconomic dimensions of cities, as they primarily encode physical and spectral patterns rather than human activities or spatial functions. We propose AETHER(AlphaEarth–POI Enriched Representation Learning), a lightweight framework that adapts AlphaEarth to human-centered urban analysis through multimodal alignment guided by Points of Interest (POIs). AETHER aligns AE embeddings with textual representations of POIs, enriching physically grounded EO features with semantic cues about urban functions and socioeconomic contexts. In Greater London, AETHER achieves consistent gains over the AE baseline, with a 7.2% relative improvement in land-use classification F1 and a 23.6% relative reduction in Kullback–Leibler divergence for socioeconomic mapping. Built upon pretrained AE, AETHER leverages a lightweight multimodal alignment to enrich it with human-centered semantics while remaining computationally efficient and scalable for urban applications. By coupling EO with human-centered semantics, it advances geospatial foundation models toward general-purpose urban representations that integrate both physical form and functional meaning.

This work demonstrates the potential of foundation-model-encoded satellite embeddings to bridge Earth's surface and atmosphere. Based on our multimodal foundation model Copernicus-FM, we curate a global embedding dataset at 0.25°x0.25° resolutions (in consistency with ERA5). For each grid, multi-sensor images from Sentinel-1, 2, 3, 5P, and DEM are encoded into an embedding vector. These grid embeddings serve as condensed surface representations for downstream users. We verify their benefits as input for a climate task that predicts the 10-year mean and standard deviation of several climate parameters (e.g., precipitation) from ERA5. Compared to raw coordinates or location encodings, our results suggest that introducing surface embeddings helps produce more accurate prediction maps, reducing RMSEs by an average of up to 45%.

Ground-level horizon imagery is scarce across planetary bodies, making representation-centred approaches attractive for downstream geospatial tasks. We present DESPINA, a geospatial representation system that converts digital elevation models (DEMs) into structured neural embeddings of terrain geometry that condition a diffusion model to produce geometry-preserving, terrain-consistent visual reconstructions for a specified location and view direction.

Our pipeline integrates numeric elevation data (DEMs), structural embeddings (inverse-depth and soft edges), and textual priors, unifying heterogeneous geospatial signals into a shared, metric conditioning space. Using a Stable Diffusion model constrained with ControlNet, we can generate geologically consistent yet texturally diverse horizon datasets. Appearance priors are learned from historical surface photography to capture realistic textures and lighting cues, and geometric validation is performed against DEM-derived skylines and depth structure, independent of photographic training data. Through quantitative evaluation and a pilot qualitative study, DESPINA maintains skyline fidelity and geological boundaries while improving structural similarity relative to an image-conditioned baseline. Although our experiments use lunar DEMs and historical surface photography, the method is domain-agnostic and applicable to Earth, Mars, and other planetary DEMs.

Satellite -based crop classification and maps are important tools for food security and climate change mitigation, but existing approaches are not effective for small field systems. To address this, crop type classification using embeddings generated by a global foundation model, TESSERA, are compared to standard classification approaches in the literature. We find that our embedding -based approach offers a triple win: 1) consistent and statistically significant performance improvement over current methods, 2) greater simplicity due to the elimination of feature engineering, and 3) the reduction of computational cost. Our embedding -based approach achieves significantly higher F1 scores in the classification of 5 of 7 crop types for small fields in Austria (over 10% improvement in one case). Additionally, the TESSERA embedding -based method uses 8% of compute compared to the raw data method. These results indicate that embeddings are an effective approach for crop type classification tasks in small field systems.

This study explores the potential of the AlphaEarth Foundation (AEF) embeddings, a global, annual, analysis-ready satellite embedding dataset, for winter wheat and crop rotation mapping. Firstly, we analyze AEF embeddings for intra-class consistency and inter-class separability, assessing their effectiveness in representing wheat within the semantic embedding space. Subsequently, we compare multiple lightweight classifiers to identify an optimal model and conduct spatiotemporal generalization experiments across Henan Province from 2018 to 2024 using only a limited set of labelled samples from 2020. Based on the resulting wheat distribution maps, crop rotation patterns are further identified.Experimental results demonstrate that AEF embeddings exhibit strong semantic coherence and discriminative capability. Acceptable classification accuracy (OA = 0.85) can already be achieved using simple models such as cosine similarity and linear regression. More advanced lightweight classifiers further improve the performance (OA = 0.86–0.93) while maintaining stable results across different years and regions (spatial consistency = 0.82). In addition, the crop rotation maps show high spatial agreement with existing products, while producing more spatially contiguous field patterns.Overall, this study confirms that AEF embeddings can serve as effective, ready-to-use features for large-scale agricultural remote sensing applications. By substantially reducing the reliance on complex feature engineering and extensive training samples, they provide a practical and scalable solution for mapping winter wheat and its crop rotation patterns.

Identifying a distressed vessel among many ships detected in wide-area aerial imagery is a critical challenge in maritime Search and Rescue (SAR) operations. Conventional methods cannot determine which vessel matches the incident description, especially when Automatic Identification System (AIS) reports are uncertain. This study proposes an integrated framework that combines MLLM-based semantic attribute extraction with geolocation fusion to prioritize candidate vessels according to their consistency with Situation Report (SITREP) based scenarios. The method detects vessels using YOLOv8, tracks them with Deep Simple Online and Real-time Tracking (DeepSORT), and performs image-based georeferencing using onboard metadata. A Multi-modal Large Language Model (MLLM) extracts appearance/status attributes from representative vessel images, while scenario descriptions are also converted to attributes. Both sets are encoded using MiniLM embeddings. Finally, semantic similarity is fused with geolocation proximity within an Support Vector Machine (SVM) classifier to produce a probability-ranked list of candidates. Experiments using real aerial search footage demonstrate robust identification performance across a range of scenario quality levels. The correct vessel appears within the top three candidates in more than 73\% of cases and within the top five in more than 91\%, even when attribute extraction is affected by low resolution, illumination effects, or missing scenario information. These results show that coarse semantic cues, when combined with approximate geolocation, provide a resilient basis for identifying target vessels under high uncertainty. The proposed framework offers a practical foundation for automated SAR decision support, enabling faster and more reliable prioritization during wide-area maritime search operations.

Cross-view geolocalization (CVGL) systems, while effective at retrieving a list of relevant candidates (high Recall@k), often fail to identify the single best match (low Top-1 accuracy). This work investigates the use of zero-shot Vision-Language Models (VLMs) as rerankers to address this gap. We propose a two-stage framework: state-of-the-art (SOTA) retrieval followed by VLM reranking. We systematically compare two strategies: (1) Pointwise (scoring candidates individually) and (2) Pairwise (comparing candidates relatively). Experiments on the VIGOR dataset show a clear divergence: all pointwise methods cause a catastrophic drop in performance. In contrast, a pairwise comparison strategy using LLaVA improves Top-1 accuracy over the strong retrieval baseline. Our analysis concludes that VLMs are poorly calibrated for absolute relevance scoring but are effective at fine-grained relative visual judgment, making pairwise reranking a promising direction for enhancing CVGL precision.

Accurate and topologically consistent vector data extraction from aerial imagery is essential for geospatial applications such as urban mapping and the enrichment of volunteered geographic information (VGI) platforms like OpenStreetMap (OSM). While deep learning has advanced automated feature extraction, traditional supervised networks remain constrained by large annotation requirements and limited generalization. Recent vision foundation models offer a promising alternative through promptable, zero-shot segmentation capabilities.

This study presents a modular “pixels-to-polygons” workflow for transforming high-resolution orthophotos (20 cm DOP20) into GIS-ready vector data, comprising four stages: data preparation, segmentation, vectorization, and validation. The framework is model-agnostic and designed to integrate different vision foundation models without modification to downstream processing. As a representative instantiation, we evaluate the Segment Anything Model 2 (SAM2) in automatic mask generation mode over the University of the Bundeswehr Munich campus.

The resulting segmentation masks are polygonized and compared against OSM reference layers. Preliminary results indicate that vision foundation models can effectively delineate major built-up areas, vegetation, and transport infrastructure without task-specific training, producing geometrically smooth and topologically valid polygons. However, smaller or shadowed objects remain challenging, often leading to partial merging or fragmentation.

Future work will include quantitative evaluation using raster-based Intersection-over-Union and vector-domain metrics such as completeness, correctness, and geometric fidelity. The study provides an initial assessment of vision foundation models for high-resolution remote sensing and highlights their potential to bridge the gap between general-purpose segmentation models and GIS-compatible vector data generation.

Surface-laid unexploded ordnance (UXO) and landmines constitute a critical humanitarian crisis. While unmanned aerial vehicles (UAVs) provide a scalable remote sensing solution, detecting modern, non-metallic explosive devices in cluttered environments remains a profound Camouflaged Object Detection (COD) challenge. Traditional optical sensors frequently suffer from foreground-background confusion when a target's texture mimics its surroundings. To overcome these physical bottlenecks, we introduce XPol-Net, a novel multimodal architecture synergizing the semantic reasoning of Vision Transformers with the deterministic physics of polarimetric imaging. Built on a hierarchical PVTv2 backbone, XPol-Net utilizes a progressive Dual Cross-Attention Strategy for effective modality fusion. In early stages, Channel Cross-Attention (CCA) filters material-specific Degree of Linear Polarization (DoLP) cues to suppress background clutter. In deeper stages, Spatial Cross-Attention (SCA) dynamically aligns high-level RGB semantics with strict structural boundaries. To enhance robustness and prevent modality collapse, we deploy a multi-task auxiliary learning framework that reconstructs the continuous Angle of Linear Polarization (AoLP) map. On the PCOD benchmark, XPol-Net achieves state-of-the-art results in global structural alignment (E_phi of 0.980 and 0.984 at 352 x 352 and 704 x 704, respectively). While minor trade-offs are observed in localized metrics such as S_alpha or F_beta, XPol-Net remains highly competitive, consistently delivering superior results in E_phi and MAE. By prioritizing structural recall over localized strictness, XPol-Net ensures the complete discovery of concealed targets, establishing a reliable, physics-aware foundation for humanitarian demining operations.

A novel hand–eye calibration method for ground-observing mobile robots is proposed. While cameras on mobile robots are common, they are rarely used for ground-observing measurement tasks. Laser trackers are increasingly used in robotics for precise localization. A referencing plate is designed to combine the two measurement modalities of laser-tracker 3D metrology and camera-based 2D imaging. It incorporates reflector nests for pose acquisition using a laser tracker and a camera calibration target that is observed by the robot-mounted camera. The procedure comprises estimating the plate pose, the plate–-camera pose, and the robot pose, followed by computing the robot–-camera transformation. Experiments indicate sub-millimeter repeatability.

Early detection of structural issues is crucial for timely maintenance and extending bridge lifespans. Conventional visual inspections alone are not sufficient for the vast number of structures, highlighting the need for intelligent, real-time monitoring systems. The studies presented here were conducted as part of the DFG priority programme SPP100+ in collaboration with the IDA-KI project. In addition to conventional civil engineering sensors, the 45-meter-long openLAB research bridge, a three-span prestressed concrete structure, was equipped with target fields for photogrammetric measurements. During controlled load tests, in which a motorized load vehicle passes over the bridge, a high-resolution camera captures image sequences of the measurement fields. The photogrammetric workflow involved camera calibration, image sequence acquisition, and precise 3D coordinate determination using coded targets. Displacement values in the range of several millimeters were calculated frame-by-frame. Results from a typical load cycle showed initial upward deflection followed by downward movement, corresponding to the load vehicle’s passage. This approach demonstrates the potential of photogrammetry for accurate, non-contact deformation monitoring, supporting the development of digital twins and advanced structural health monitoring systems for bridges.

Body donation remains critically important for anatomical science, allowing examination of biological structures with three-dimensional (3D) context. However, body donors (cadavers) are a time-limited resource and the scarcity of body donors has prompted an interest in digital body preservation. Multiple imaging techniques (e.g., conventional photogrammetry[CPG], stereophotogrammetry[SPG] and computed tomography[3DCT]) can capture the 3D characteristics of a specimen indefinitely. Digital anatomical records provide an opportunity to measure anatomical structures in the absence of the physical specimen. In 2022, the face of a preserved body donor was digitally reconstructed using CPG and 3DCT. 28 months later, a repeat survey was performed using SPG and a series of facial landmarks were directly measured. The accuracy and stability of facial soft-tissues over time were measured using point-to-point and cloud-to-mesh techniques. The results show that anatomical models produced by 3DCT and CPG produce similar facial measurements to those acquired by SPG and direct measurement at later timepoints. These data indicate that chemical fixation adequately stabilises facial anatomy over time, each sensor can be used interchangeably for facial measurement and models can be co-registered with minimal discrepancy.

Speech is a highly complex and multidimensional process, requiring precise coordination of muscular actions within the vocal tract. Disruptions or delays in speech motor control often lead to speech impairments. Recent advancements in markerless facial tracking technology enable the collection of objective measurements to assess these impairments. To obtain such photogrammetric measurements, a multi-camera network is employed, making accurate camera calibration essential. This paper examines the constraints applied during the calibration process. Two adjustment strategies were evaluated. The first, Independent Adjustment (IDP), performs self-calibration for each camera without introducing constraints. The second, Combined Adjustment (CMB), incorporates object space constraints by ensuring that object point locations observed from all cameras remain consistent. Given the cameras’ narrow fields of view, both IDP and CMB were tested with additional constraints related to the principal point offset. Each adjustment was executed under two conditions: fixing the principal point offset to zero or estimating it as part of the calibration. Results indicate that the choice of adjustment significantly affects the interior orientation parameters (IOPs). IDP with the principal point offset fixed to zero produced the most accurate outcomes. However, variations in IOPs had no meaningful impact on object space coordinates. These findings suggest that the simplest approach—IDP with the principal point offset fixed to zero—offers reliable calibration for multi-camera systems used in speech assessment. This streamlined method can be adopted in future applications to enhance efficiency without compromising accuracy.

Daily satellite-derived nighttime lights (NTL) are increasingly used to monitor electricity disruption and recovery, but their reliability in tropical regions is constrained by persistent cloud cover and intermittent observation. This study adopts a diagnostic-first framework that treats observability as a prerequisite for interpretation, determining when daily NTL signals are sufficiently supported to reflect underlying grid dynamics. Using NASA’s VIIRS Black Marble product, we quantify spatial and temporal completeness and radiance stability across the Samar–Leyte sub-grid in the Philippines. Results show that observations are highly intermittent and spatially heterogeneous, with urban areas providing more stable and interpretable signals, while rural regions remain noise-dominated.

To assess whether reliability-qualified NTL reflects electricity demand, DNB-BRDF radiance is aligned with hourly load data from the National Grid Corporation of the Philippines (NGCP) using settlement-based masks derived from GHSL SMOD. Alignment is evaluated using correlation, error, and retained coverage, combined into a composite score. Strongest agreement occurs under low to mid-range valid-pixel thresholds and within urban-focused masks, which balance signal fidelity and temporal continuity at the cost of reduced coverage. Replication across four additional Visayas sub-grids shows that optimal threshold–mask configurations vary by region, reflecting differences in cloud regime and settlement structure.

These results establish explicit conditions under which daily NTL can be interpreted as a proxy for grid dynamics. The framework provides a reproducible basis for reliability-qualified analysis using globally available datasets and can be tested in other cloud-prone regions where ground-based data are limited.

The maintenance of airport pavements is critical to ensuring the safety and efficiency of air operations. Conventional inspection methods are often time-consuming, subjective, and prone to inconsistencies in data collection. Recent advances in unmanned aerial vehicle (UAV) photogrammetry offer a potential alternative for improving inspection efficiency and measurement accuracy.

This study evaluates the applicability of UAV-based photogrammetry for the detection and quantification of pavement distresses under conditions representative of airport infrastructure. Image data were acquired at different flight altitudes and overlap configurations and processed using Structure-from-Motion techniques to generate high-resolution orthomosaics and Digital Elevation Models (DEMs). The resulting datasets were analyzed to identify, delineate, and classify deterioration types and severity levels.

The results indicate that a flight altitude of 10 m combined with 80% longitudinal and 70% transversal overlap provides an optimal balance between spatial resolution and operational efficiency. Under unobstructed conditions, photogrammetric analysis detected more than 98% of existing distresses and enabled more precise geometric delineation compared to traditional field-based methods. Undetected distresses were primarily associated with shadowed or obstructed areas, highlighting the influence of environmental conditions on detection performance.

Overall, the findings demonstrate that UAV-based photogrammetry is a reliable and efficient approach for pavement condition assessment, with significant potential to enhance data quality and reduce inspection time in airport infrastructure management.

3D change detection is essential for monitoring infrastructure, environmental dynamics, and natural hazards. However, existing algorithms are often evaluated on single-scene datasets, and their generalization across varied real-world scenes remains largely unexplored due to the absence of a universal benchmark. To address this issue, we propose MultiChange3D, a multi-scene, multi-sensor 3D change detection dataset for identifying geometric changes in 3D space. The dataset provides registered pairs of point clouds with ground-truth geometric change labels, enabling standardized evaluation across different methods. To demonstrate the use of the MultiChange3D dataset, we benchmark an initial set of approaches on a subset of the dataset. The evaluated methods include classical Euclidean distance-based methods (C2C, M3C2), 3D displacement estimation-based approaches (F2S3, Landslide-3D), and deep learning-based classification methods (KPConv, EF-KPConv, PGN3DCD). Quantitative and qualitative analyses indicate the strengths and limitations of the evaluated methods, highlighting the challenges in cross-scene generalization under variations in point density, scale, and types of changes. The full dataset and evaluation code is openly available at: https://github.com/3DOM-FBK/multichange3d.

Quantifying topographic dynamics from 3D point cloud time series is essential for geoscientific applications. However, laser scanning data typically varies between epochs in point density due to differing survey properties. These irregularities present a challenge for change detection, particularly across multi-temporal and multi-modal data. We propose a new approach, adaptive temporal aggregation, as an extension of the Multiscale Model to Model Cloud Comparison (M3C2) algorithm. Driven by a local point density requirement, our method employs both a spatial and a temporal neighborhood. If a core point's neighborhood is too sparse for M3C2 estimation, an iterative temporal search progressively incorporates data from temporally adjacent epochs until the density requirement is met or a maximum temporal window is reached. This adaptive process ensures sufficient local density while preventing unnecessary temporal aggregation, a key advantage over global aggregation. We evaluated our method on a multi-modal dataset from the Mont-Saint-Michel Bay, France (38 irregular epochs, ~1 decade). Results demonstrate significantly improved change detection, increasing completeness by >13% (vs. standard M3C2) and accuracy by 31% (vs. fixed-window averaging). Our work provides a robust approach for enhancing 3D change detection algorithms for complex, real-world 4D datasets, enabling higher accuracy and completeness in analysing surface dynamics.

This study evaluates the classification performance of Sentinel-2 multispectral and EnMAP hyperspectral imagery across three Canadian agricultural sites. Using Random Forest with recursive feature elimination, we assess whether EnMAP’s spectral richness provides measurable improvements for operational crop mapping. Results show comparable overall accuracies, with EnMAP offering advantages for spectrally complex crop types. Findings highlight the potential and practical limitations of incorporating satellite hyperspectral data into national agricultural monitoring workflows.

Agricultural land in Slovakia has undergone a significant transformation over the past two decades, but the extent and direction of this change remain insufficiently quantified at the national level. This study provides the first spatially explicit assessment of agricultural land loss using Land Parcel Identification System (LPIS) records (2004–2022) in combination with Sentinel-2-based land cover classification. By differentiating the LPIS, we identified polygons that were lost from the agricultural land register; these polygons could represent abandoned agricultural land as well as areas converted to other uses. These polygons were classified into four land cover classes using a Random Forest model trained on Sentinel-2 spectro-temporal metrics and cleaned LUCAS 2022 samples (F1 = 0.867), with additional filtering applied to separate grasslands from shrubs based on vegetation height. The results show that more than 1,000 km² of originally suitable agricultural land has been converted to other land cover types. Forest expansion accounts for 834 km², while 298 km² has been converted to shrubland and 553 km² remains as grassland. Non-forested areas, including buildings and infrastructure, cover an area of 258 km². Only 17 km² of formerly agricultural land remained as actively utilised arable land. These findings indicate that agricultural land abandonment, ecological succession, and urbanisation are the primary causes of agricultural land loss in Slovakia. The research presented provides important data confirming that the loss of agricultural land is extensive and largely threatens habitats with high biodiversity, highlighting the urgent need to harmonise strategies across agriculture, the environment, and land-use planning.

Agricultural drought affects crop productivity and threatens food security. This study presents the Hierarchical Robust Combined Drought Index (HRCDI) for operational agricultural drought monitoring, based on Earth Observation (EO) data freely available in Google Earth Engine. The HRCDI integrates four complementary indicators—Standardized Precipitation Evapotranspiration Index (SPEI3), Soil Moisture (SM), Land Surface Temperature (LST), and Normalized Difference Vegetation Index (NDVI)—using a hierarchical fuzzy logic framework that reflects the progression of drought impacts from climatic anomalies to vegetation stress. To ensure robustness, monthly anomalies of SM, LST, and NDVI were computed through a robust z-score formulation based on median and NMAD, which reduces the influence of outliers. The HRCDI was applied to the Province of Foggia (southern Italy), one of the main durum wheat production areas in Italy, over the period 2017–2022. HRCDI outputs were aggregated at the municipality scale and validated against independent datasets, including durum wheat yield statistics (2006–2022), SPEI3 provided by the Institute of Services for Agricultural and Food Market (ISMEA), and reports from the European Drought Impact Database. The HRCDI effectively captured the severity and spatial extent of major drought events, particularly in 2017 and 2022, which corresponded to documented yield losses of −5% and −22%, respectively. Results highlight the scalability, operational relevance, and transferability of the HRCDI for supporting drought early warning and agricultural risk management. The HRCDI framework could be applied to other regions and integrated with higher-resolution satellite data to enhance drought monitoring in line with the objectives of the Common Agricultural Policy.

Fall Armyworm (FAW) is a major pest, threatening maize production and food security. FAW preferentially attacks nitrogen-rich maize due to its nutritional composition, which accelerates larval development. Understanding the effects of FAW infestation on maize growth and nutrient status is critical for effective crop management. This study (i) examines the impact of FAW infestation on maize using key physiological metrics, including fresh/dry weight, chlorophyll content (as measured by SPAD), leaf area index (LAI), stem length, and leaf dimensions (length, width, and area); (ii) investigates differences in carbon, hydrogen, nitrogen, and sulphur concentrations between healthy and infested maize, and (iii) analyses spectral variability between healthy and infested maize using leaf-level hyperspectral data. Hyperspectral data at the leaf scale were used to identify and distinguish the spectral reflectance between healthy and infested FAW maize crops. Results indicate that infested crops exhibit lower fresh weight, reduced LAI, shorter stems, and smaller leaves compared to healthy crops, highlighting a substantial negative effect on above-ground biomass and overall crop vigour. Infested crops showed higher nitrogen levels than healthy crops. This trend could be attributed to nitrogen redistribution following FAW damage, where nitrogen decreases in attacked leaves and increases in roots, with partial recovery after the pest moves on to another crop. The study establishes a methodological framework for linking laboratory, field, and remote sensing approaches, providing a foundation for future predictive modelling of pest impacts on maize nutrient content and productivity.

Urban digital twin systems require 3D city representations that reconcile semantic structure, geometric reliability, simulation capability, and photorealistic real-time rendering. Existing approaches typically prioritize a single modelling paradigm, limiting their capacity to simultaneously support analytical and visualization demands. CityGML ensures standardized semantics and topological consistency but often lacks detailed surface realism. Surface-based semantic mesh models preserve geometric detail suitable for environmental simulations but provide limited hierarchical semantic organization. In contrast, neural radiance-field approaches, such as 3D Gaussian Splatting (3DGS), enable photorealistic rendering at interactive frame rates but do not explicitly encode topology or structured semantics. This study establishes a structured comparative framework linking LiDAR-derived CityGML, triangle-based semantic mesh, 3D Gaussian Splatting, and Triangle Splatting within a unified urban modelling workflow. UAV-based data acquired using a multirotor platform with a DJI ZENMUSE L2 sensor serve as the geometric backbone for reconstructing CityGML LoD1-LoD2 models. The semantic model is transformed into a textured triangular mesh to provide a geometry-consistent baseline, while radiance-based models are generated from the same imagery using multiple 3DGS implementations and a triangle splatting framework. Comparative evaluation investigates geometric coherence, semantic preservation, and radiance consistency to identify structural correspondences across the representations. Rather than treating them as competing alternatives, the results reveal complementary modelling layers that can be systematically mapped. Based on these findings, the paper formulates a conceptual foundation for a unified 3D urban model capable of transforming consistently into semantic-structured, surface-based, and radiance-based representations, enabling adaptive and extensible urban digital twin systems.

Persistent Scatterer Interferometry (PSI) provides valuable information on ground and structural changes, particularly in dynamic urban environments. At the same time, urban digital twins (UDTs), as detailed three-dimensional representations of cities, are increasingly used for monitoring and analysis. However, the effective integration of results of PSI processing named Point Scatterers (PSs) into such frameworks remains challenging due to the limited positional accuracy of PSs, despite the high precision of displacement estimates. This study investigates a methodology for integrating PSI data from the European Ground Motion Service (EGMS) with airborne laser scanning (ALS) data and Level of Detail 2 (LoD2) building models to improve the connection of PSs with real-world objects. Three integration variants were analysed, differing in the reference datasets used for linking: (i) ALS point cloud, (ii) point cloud derived from LoD2 models and digital terrain model (DTM), and (iii) a combined approach integrating ALS and LoD2 representations. The results demonstrate that the combined approach yields the highest performance, achieving up to 88% of successfully linked PSs, compared to 70.4% and 80.3% for the ALS-only and LoD2-based approaches, respectively. The findings indicate that LoD2 models provide sufficient geometric detail for PS linking, despite lacking fine-scale building elements. Their use improves data completeness, particularly on building facades, where ALS data are often sparse or missing. The proposed methodology confirms the applicability of EGMS products as a valuable data source for 3D geoportals and urban digital twins, supporting advanced spatial analyses in complex environments.

Integrated Water Service (IWS), which combines water supply and wastewater treatment, requires complex geometric and semantic management. Building Information Modelling (BIM) and Geographic Information Systems (GIS) are the two main geospatial technologies involved in this field. In very simple terms, BIM allows to have 3D models with detailed geometric and semantic information, and GIS permits to geolocate and manage the models in the territory. To facilitate the integration of these two systems, we propose to manage the BIM models through a standardised relational database. In the BIM world, relational databases are not yet widely used, but the technology is already available. For example, ifcSQL is an encoding of the Industry Foundation Classes (IFC) data model for a relational database. This article proposes an extension of the ifcSQL database with the added possibility to store the georeferenced explicit geometries of the IFC models. Additionally, we present a prototype to make such IFC-based data available via QGIS. In this way, a user can interact with BIM data using open GIS technologies. As a result, it is possible to visualise the models in 2D and 3D, and to perform queries on their attributes. A set of real-world case studies has served as testing ground to develop the functionalities that allow for the interaction with the BIM models via QGIS. Such test cases originate from interviews with a company that manages IWS in Northeast Italy.

The increasing demand for safe and efficient autonomous vehicle (AV) operations has intensified the need for realistic, high-fidelity digital road representations that enable robust virtual testing environments. Simulation-based validation has become a cornerstone of the AV development process, allowing for the reproducible assessment of perception, localization, and decision-making modules under controlled conditions. Within this context, the ASAM OpenDRIVE specification provides a standardized, XML-based description of static road networks, encapsulating geometric, semantic, and structural elements such as roads, lanes, junctions, and roadside objects. While previous research has primarily focused on the geometric accuracy and semantic richness of High Definition (HD) maps, comprehensive topological analyses—especially those addressing consistency, connectivity, and completeness of OpenDRIVE models—remain largely unexplored. This study aims to fill that gap by introducing a formal topological framework for evaluating OpenDRIVE-based road models through both synthetic and real-world test cases.

This paper presents a comprehensive toolkit for semi-automated High-Definition Maps (HD Maps) production that integrates Artificial Intelligence (AI)-driven feature extraction with 3D human-in-the-loop validation. High-definition maps provide centimeter-level road geometry and traffic asset information, but large-scale production remains costly due to dense mobile mapping data and manual digitization. The proposed workflow consists of two self-developed components: a Semi-automated HD Maps Production Tool for batch extraction and a 3D HD Maps Validation and Editing Tool for structured review. The project-based pipeline ingests georeferenced mobile laser scanning point clouds, Inertial Navigation System / Global Navigation Satellite System (INS/GNSS) trajectories, and camera imagery, and applies configurable chains of ground filtering, road-marking extraction, voxel down-sampling, clustering, oriented bounding box analysis, and AI-based traffic asset detection. Candidate features with confidence indicators and basic attributes are stored in a project database and edited in a tightly coupled 3D environment that supports snapping, constrained adjustments, and semantic reclassification while logging all user edits. The toolkit is evaluated on a closed proving ground (CARLab, Shalun) and a freeway section of Taiwan National Highway No. 3. At CARLab, semi-automated extraction achieves F1-scores of 0.85–0.95 for key layers. For a one-kilometer highway section, operator time is reduced from 90–120 minutes in a purely manual Geographic Information System (GIS) workflow to about 45 minutes with the proposed approach, while maintaining comparable geometric accuracy. These results demonstrate a practical path towards scalable, traceable HD Maps production for autonomous driving applications.

National Key Research and Development Program of China (2025YFB3910300);

3D scene reconstruction lies at the core of computer vision, photogrammetry, and geospatial science, spatial intelligence, aiming for accurate, photorealistic, and efficient digital twin representations of the real world. The emergence of revolutionary 3D Gaussian Splatting (3DGS) enables real-time rendering and geometrically precise reconstruction, yet existing methods struggle in large-scale outdoor scenes with weak textures, low geometric accuracy, dynamic objects, and lack of semantic information. Therefore, geometrically accurate 3D GS with enhanced semantic understanding greatly facilities the realization of digital twins for mobility and navigation. This work proposes a novel 3DGS framework which seamlessly incorporates dense UAV LiDAR point clouds, multi-view images and open-set semantics in an all-in-one optimization process. The key objective here is to investigate how geometric constraints derived from dense UAV LiDAR point clouds and cognitive supervision from SAM (Segment Anything Model) semantics can jointly participate in the optimization of Gaussian primitives, thereby improving geometry accuracy, visual realism, and semantic consistency in large-scale UAV 3D reconstructions for creating digital twins of the environments.

The restoration of historic buildings is an important topic in today's society and constitutes the primary subject of this study. The Diji Altar, located along the central axis of Beijing, is not only a significant historical landmark but also an important remnant of China's ancient imperial sacrificial architecture. Although some studies have focused on the Diqi Altar, such as its ritual hierarchy and craftsmanship as recorded in historical texts, certain research gaps remain. Due to the damage to the altar structure and insufficient documentation in relevant literature regarding its structural form, platform base specifications, and stylistic evidence, systematic research on restoration techniques remains relatively scarce. There is a need to reconstruct evidence based on architectural principles. Addressing this critical gap is of great significance for understanding the technical achievements and ceremonial principles of official architecture during the Ming and Qing dynasties, and for guiding the restoration and preservation of ancient buildings.

The digital preservation of open-air sites often faces multiple challenges, such as dispersed components, varied forms, and missing historical records. In response, this study focuses on the Beijing Dizhitan and proposes and implements an innovative workflow that deeply integrates architectural morphology theories, archaeological typological methods, and modern digital technologies. This workflow systematically constructs a complete methodological chain, from the semantic annotation, classification, and virtual assembly of stone components, to the virtual restoration and model reconstruction of the site, ultimately achieving scenario-level restoration and display evaluation. The successful restoration of the Dizhitan demonstrates that this approach not only effectively "revives" dispersed components, placing them in their proper positions in a virtual space, but also pioneers a replicable new paradigm that embeds rigorous academic research throughout the digital process. This provides an entirely new technical approach and perspective for the preservation, study, and interpretation of immovable open-air cultural relics.

This paper addresses the limitations of single-modal data in rock art cultural heritage preservation, such as incomplete information and fragmented semantics. It proposes a method for constructing a multimodal dataset that integrates text, images, and 3D point clouds. Text data is structured and semantically annotated using the ArchaeoBERT model; image data is obtained through web scraping, annotation, and augmentation; and point cloud data is captured using laser scanning, noise reduction, and registration techniques. Feature mapping alignment is employed, combining CNN, BERT, and PointNet++ to extract features and generate unified vector representations. Through a three-level quality control process, the data is accurate and reliable, with information coverage increased by 47.3%. This dataset achieves comprehensive integration of semantic, visual, and spatial information, providing a multidimensional data foundation and practical reference for the digital preservation, 3D reconstruction, and cross-modal retrieval of rock art.

The rapid evolution of Monocular Depth Estimation (MDE) models — and in particular the emergence of recent foundation models such as Depth Anything v2 (Yang et al., 2024; Ranftl et al., 2022) — is opening concrete perspectives for the application of artificial intelligence in architectural and cultural heritage surveying.

This research aims to assess the feasibility of employing such models to obtain metric depth estimations from UAV imagery, acquired in both oblique and nadir views, with the broader goal of integrating neural networks into 3D documentation, HBIM, and GIS workflows for built heritage.

The Depth Anything v2 models were trained initially for ground-level scenarios, where the camera typically operates 1–2 m above the ground, with horizon distances extending up to 60–80 m. When applied to aerial imagery, particularly drone-based acquisitions, this results in a substantial domain gap: the network tends to interpret top-down landscapes as distant horizons, thereby compressing the depth scale.

To address this issue, this study develops an experimental calibration and adaptation procedure aimed at transforming the depth maps produced by the model into metrically consistent estimates that are coherent with architectural reality.

We present an approach to ensure trustworthy, high-quality GeoAI which requires a coordinated effort across academia, industry, government, and standards bodies. The ISPRS community, in partnership with organizations such as ISO, OGC, the World Geospatial Industry Council (WGIC), the International Society for Digital Earth (ISDE), and other international initiatives is uniquely positioned to host this dialog. By aligning emerging GeoAI practices with established data-quality standards and ethical-AI frameworks, the community can help shape a future-proof foundation for responsible innovation in geospatial intelligence.

This paper presents research on integrating Data Quality and Licensing metadata into Open Science workflows using ISO and OGC standards and machine-readable profiles to enhance interoperability, transparency, and reusability of scientific data. All if this is possible via ongoing development of modular building blocks, validation frameworks, and engagement with standards bodies to support FAIR principles and scalable data reuse across domains. Our approach demonstrates the possible integration of concept schemes and measures defined in the ISO 19157 multipart standard.

This paper presents the ISO 19157-3 Data quality measures register, discusses its design and implementation and illustrates its the utility to the ISPRS and wider geocommunity. In this paper we highlight the importance of providing geographic metadata about quality, the evolution of international standards to support this, and a novel implementation of a human readable and machine-actionable web register for geographic data quality measures.

The rapid growth of geospatial data, fueled by advancements in satellite imagery, IoT sensors, and mobile services, presents significant opportunities in sectors like urban planning and environmental monitoring. However, these data are also vulnerable to cyber threats, emphasizing the need for strong protection mechanisms. This paper introduces a modular, hybrid architecture that addresses security challenges by integrating post-quantum cryptography, decentralized storage, and access control via Blockchain. It employs AES-GCM for the secure encryption of large datasets and Kyber for enhanced key protection against quantum threats. Encrypted data is stored securely in the Interplanetary File System (IPFS), with access managed by smart contracts on a private Ethereum blockchain. The architecture utilizes FastAPI for back-end processes, microservices for cryptographic services, and React for the user interface. Performance assessments show good scalability and resilience, paving the way for secure geospatial data sharing while harmonizing data sovereignty, quantum security, and decentralized management.

High-Resolution Global Land Cover (HRLC) products are essential for monitoring Earth’s surface dynamics and supporting policy frameworks like the Sustainable Development Goals. Recent global products such as ESA WorldCover, ESRI LULC, FROM-GLC, and Dynamic World offer 10–30 m resolution maps, but their interoperability remains limited due to differences in input data, class legends, and validation protocols. This lack of harmonization hampers cross-comparison and integrated use for environmental monitoring. Although advances in remote sensing, AI, and cloud computing have enabled more frequent and detailed mapping, they have also introduced new challenges for ensuring data consistency and comparability.

Validation of HRLC products is hindered by the absence of a common benchmark dataset, as current accuracy metrics are derived from heterogeneous reference samples and class definitions. Traditional validation methods are costly and time consuming, while temporal inconsistencies and cloud contamination further increase uncertainty. ISO 19157-3 offers a standardized framework to describe and automate quality measures such as positional accuracy and thematic correctness, supporting transparent and reproducible evaluation across datasets.

A sustainable solution involves establishing an international benchmarking framework with standardized reference data, legends, and sampling strategies. As a practical interim approach, the Map of Land Cover Agreement (MOLCA) combines multiple HRLC products to identify spatial consensus and disagreement, offering a proxy for thematic reliability. Although MOLCA measures consistency rather than absolute accuracy, its integration into ISO 19157-3 would advance data quality assessment, fostering transparency, interoperability, and confidence in HRLC-derived environmental analyses.

The goal of the CO3D (Constellation Optique 3D) mission is the full-automatic production of a worldwide accurate DSM (Digital Surface Model). This DSM is generated from stereo acquisitions obtained from a new generation of high-resolution optical satellites, called CO3D. The DSM accuracy is one meter in relative height for moderate slopes and four meters in absolute height with a one-meter grid space. Each of the four satellites of the constellation provides images with 0.50 m resolution in red, green, blue and near-infrared bands. The satellites resource is shared by, on one hand, the French institutions (government, scientists concerned by global Earth monitoring) who have dedicated access and preferred price conditions, and on the other hand, ADS (Airbus Defence and Space) customers interested in 2D and 3D products. The constellation was launched on July 26, 2025

We present the qualification of 3D products as part of the CO3D mission. The CO3D mission is dedicated to creating a digital surface model of the Earth's landmass cover. Massive automatic production is a challenge in itself, as this ground segment produces more advanced data for an optical mission. Firstly, the 3D product, which is generally retouched and checked, which can represent a significant cost. In the case of the CO3D mission, these products will be generated completely automatically. Masks will also be included to describe the processing history and provide precise information on the altitudes restored. All of this data requires detailed qualification with precise reference data and methods to best reflect its quality. The paper will describe all of these methods and data and provide an overview of the performance of these new CO3D products.

The launch of the four CO3D spacecrafts took place on 26th July 2025 aboard Vega-C from Kourou space center with Microcarb microsatellite. After a first week spent calibrating the most critical subsystems of the spacecraft, the instrument was switched on, enabling the 9-months Image Quality commissioning phase to begin. Images are acquired in RGB and NIR spectral bands with a 50 cm Ground Sampling Distance (GSD), thanks to matrix sensors based on a Bayer pattern. This brings new calibration challenges such as demosaicing and 2D line of sight determination. 3D calibration activities take place in a second stage of the commissioning phase once radiometric and geometric calibration are finalized.

This work introduces CO(ast)3D, a predictive pipeline that helps identify when upcoming CO3D satellite overpasses are likely to capture surface wave signals suitable for bathymetry acquisitions. Because CO3D cannot directly image the seafloor, depth must be inferred from the optical expression of surface waves, whose visibility depends strongly on illumination, viewing geometry, and sea state.

The pipeline combines CO3D orbital tracks with forecasted wave parameters from the Copernicus Marine model to construct a directional wave spectrum, generate a time-varying free surface through linear wave theory, and simulate CO3D-like radiance images at native spatial resolution. These synthetic scenes allow the clarity of the wave field to be evaluated a priori for any future time and location.

By predicting whether conditions will yield a sufficiently coherent wave signal, the system supports more efficient tasking, reduces acquisition risk, and improves the likelihood of capturing images suitable for accurate bathymetric inversion.

The challenges, main design elements, and results of the CODIP and ICC components will be presented in this paper, as well as the modalities for accessing and using the CO3D products.

Understanding inter-band and cross-channel relationships is essential for human color perception and object recognition. Yet, local gradients in standard convolutions are tied to fixed input–output channel pairs, and thus channels are fused by a dense, fully-coupled weight tensor: each output channel aggregates all input channels in a uniform way at every spatial location. This leads to heavy computation and does not exploit structured sparsity or selective local channel mixing. To overcome this limitation, we introduce Cube Kernel, a novel convolutional operator that introduces structured cross-channel groups into the local gradient. This design strengthens cross-channel feature fusion, improves optimization efficiency, and reduces computational overhead. Extensive building extraction experiments validate its effectiveness: Cube Kernel consistently outperforms standard convolutions and Involution when integrated into UNet, and replacing a single layer in DeepLabV3+, Swin-UNet, or UNet leads to consistent performance gains. Beyond serving as a lightweight plug-in module, Cube Kernel also scales effectively as a fundamental building block. A Cube-enhanced ConvNeXt variant, ConvNeXt-Cube, achieves state-of-the-art performance across all models (0.9095 IoU / 0.9535 F1 on WBD and 0.9133 IoU / 0.9547 F1 on WHU), demonstrating strong stackability and architectural potential. These results highlight a largely overlooked space in CNN design: enhancing cross-channel interaction at the gradient level. Cube Kernel offers a scalable and efficient alternative to deepen networks for channel mixing, laying a foundation for future advancements in convolutional architecture design.

Modeling land surface dynamics from satellite observations is crucial for revealing change patterns and predicting future states, although effective modeling methods remain limited.

For complex systems such as reaction-diffusion, two approaches have proven particularly effective:

(i) Direct modeling in the high-dimensional observation space with deep networks(e.g., (Wang et al., 2022)). These methods are often autoregressive. Errors accumulate during rolling extrapolation.

(ii) Modeling in a reduced-dimensional latent space(e.g., (Chen et al., 2022)). One reduces dimension and then learns the evolution. Some works estimate the intrinsic dimension (ID) and model in the ID latent space.

This improves long-term stability, but reliance on latent representations may reduce accuracy.

This route is promising if two issues are addressed:

(1) effectively modeling multi-scale spatiotemporal data with long sampling intervals;

(2) combining ID-space modeling with other latent dimensions to balance accuracy and stability.

This paper proposes a Dual Latent-Space Representation-based Land Surface Dynamic Model （DLS-LSDM）. The core contributions are:

(1) a stacked-convolution and multi-scale linear-attention autoencoder to obtain a base latent, together with ID estimation to derive an ID latent;

(2) a long-horizon scheme that combines ID and base latents to achieve both stability and high accuracy ;

(3) comprehensive evaluation on ten-year MODIS NDVI across multiple climate zones, demonstrating superiority.

Deep learning models have become the backbone of remote sensing image intelligent classification, enabling high-precision recognition of land cover, geospatial objects, and scene categories. However, their inherent "black-box" nature—where decision logic is embedded in complex parameter spaces—poses critical barriers to deployment in high-stakes domains such as military reconnaissance, disaster monitoring, and environmental governance. These fields demand transparent reasoning to validate model reliability, yet traditional interpretability methods suffer from two key limitations when applied to remote sensing data: They are primarily designed for natural images, failing to account for remote sensing-specific characteristics. They focus on local feature attribution or saliency mapping but lack quantitative analysis of how core image features (shape, texture, spectrum) contribute to global classification decisions, especially across different network architectures.To address these problem, this study proposes a comprehensive feature contribution analysis framework tailored to remote sensing images, with the core objectives of: (1) Decoupling and extracting shape, texture, and spectrum features from remote sensing images in a physically meaningful manner; (2) Quantifying the contribution of each feature type to classification decisions; (3) Revealing differences in feature processing mechanisms between CNN and Transformer architectures.

High-resolution optical remote sensing change detection (CD) is of great significance in urban evolution monitoring, disaster assessment, and land management. Traditional deep models often face computational, memory, and inference latency bottlenecks when processing large high-resolution imagery. To address this, we propose EfficientViM-CD: a Hidden-State Mixer based efficient remote sensing change detection network. The approach builds upon the EfficientViM backbone, migrating global interaction operations into a compact hidden state space and leveraging Hidden State Mixer based on state space duality (HSM-SSD) to fuse global context while reducing computational complexity. We employ a Siamese encoding architecture to extract multi-scale features and hidden states from paired temporal images, and utilize a Cross-Hidden Fusion module to integrate hidden semantic interactions between time points. At each scale, local difference features are computed and enhanced in hidden state space, and a multi-scale decoder reconstructs a pixel-level change probability map. We conducted experiments on four public datasets (LEVIR-CD+, WHU-CD, S2Looking, SVCD) and compared against nine state-of-the-art methods. Results demonstrate that EfficientViM-CD achieves competitive accuracy while delivering significant advantages in inference speed and memory efficiency. This method offers a lightweight, efficient, and scalable solution for high-resolution remote sensing change detection, with potential for real-time monitoring and emergency response systems.

Object detection in large, dense remote sensing imagery is difficult because targets are often small and arbitrarily oriented, and state-of-the-art detectors cannot process very large images directly without a reduction in accuracy. Tiling-based inference workflows mitigate the latter issue by running inference iteratively on overlapping tiles, but introduce pre- and postprocessing overhead for image tiling and Non-Maximum Suppression (NMS). We introduce local NMS, an asynchronous tile-wise postprocessing scheme. Local NMS runs in a separate subprocess in parallel to tile-wise inference and collects intermediate results enqueued by the inference process, immediately applying postprocessing. Intelligent reordering of tiles in a preprocessing step ensures optimal usage of computing resources. We assess our method using three state-of-the art object detection models for horizontal and oriented bounding box detection on two benchmark datasets containing large dense aerial and satellite images, DOTA-v2.0 and Izembek Lagoon Birds, stratifying by image size and average object density. Local NMS consistently reduces end-to-end runtime across models and datasets without significant impact on mAP. A maximum runtime reduction of 60.77% on large dense DOTA-v2.0 scenes could be achieved without modifying model architectures or retraining.

Recovering RAW sensor measurements from rendered sRGB images is important for radiometric calibration, low-level vision, and computational photography. However, reversing a camera’s proprietary Image Signal Pipeline (ISP) is highly challenging, especially when the ISP is unknown. Existing inverse-ISP and diffusion-based approaches have several issues: they depend on known ISPs from the sensor, require one model per sensor, or generalize poorly across camera brands.

This work presents ERD (Extended RAW-Diffusion), a unified diffusion-model framework for de-rendering sRGB images into RAW format for any given image, and does not require ISP to be known or camera information from the image. ERD extends the RAW-Diffusion architecture by incorporating camera metadata only during training, allowing the model to learn a shared representation across heterogeneous sensors. To capture global sensor characteristics, ERD introduces a conditioning mechanism, Feature-wise Linear Modulation (FiLM) for global features such as CFA patterns and color gains. To enhance structural consistency, ERD integrates a ControlNet branch that injects edge and gradient priors derived from the sRGB input, stabilizing RAW reconstruction under diverse tone-mapping operations. For practical adaptation to new sensors, ERD supports efficient few-shot tuning via LoRA.

Evaluations on Adobe FiveK (Nikon and Canon) and RAW-NOD (Nikon and Sony) show that ERD outperforms state-of-the-art baselines in PSNR and SSIM, offering improved robustness to unseen camera models. ERD enables a practical, general-purpose inverse ISP process across heterogeneous imaging devices.

Cultivated land conversion to built-up area is a core form of farmland non-agriculturalization and a major threat to farmland protection in China. Current remote sensing methods for detecting such changes face two limitations: insufficient integration of domain prior knowledge and the inability of purely data-driven models to achieve both high Precision and Recall. To address these issues, this study proposes a knowledge graph-enhanced change detection method. A multi-scale knowledge analysis framework incorporating feature, scene, and business knowledge layers is constructed to systematically integrate multi-source geographic information into structured semantic representations. A knowledge fusion residual network, KG-MS-ResNet, is designed based on ResNet-18 with modifications to the first convolutional layer for bi-temporal image inputs. TransE embeds geographic indicator knowledge into multi-scale semantic vectors, while a semantic–feature dual-path fusion strategy and a knowledge-guided attention mechanism enable deep coupling between image features and domain knowledge. Experiments in Pei County, Jiangsu Province, show that the proposed method outperforms baseline ResNet across all metrics, with Recall increasing by 4.84 percentage points and F1-score by 0.0752. The results demonstrate that integrating domain knowledge graphs with deep learning significantly improves detection performance, offering a semantically interpretable solution for monitoring cultivated land non-agriculturalization and advancing the integration of knowledge-driven and data-driven approaches in intelligent remote sensing interpretation.

Road change detection is essential for maintaining up-to-date digital maps; however, conventional update processes rely heavily on the manual interpretation of aerial imagery, leading to high labor costs and inconsistent outcomes. To address these limitations, this study proposes an automated road change detection method that integrates aerial orthophoto-based segmentation with geometric boundary deviation analysis. Road areas are first extracted from high-resolution aerial orthophotos using SegFormer, a Transformer based semantic segmentation model. The segmentation results are then converted into vector polygons for geometric analysis. Structural changes, such as newly constructed or removed roads, are detected through a difference-based comparison with historical digital maps. Simultaneously, shape changes are quantitatively analyzed by measuring geometric deviations between road boundaries.

Specifically, vertex-wise distances between corresponding boundaries are computed, and the overall deformation is evaluated using Root Mean Square Error (RMSE), incorporating Z-score-based outlier removal to ensure robustness against noise. Experimental results demonstrate that the proposed method effectively detects both structural changes and subtle geometric variations, including road expansions and boundary shifts. Furthermore, the method enables clear object-level classification of change types, providing a practical and efficient framework for digital map updating workflows.

Accurate building change detection depends on how well building boundaries are delineated, as distortions and merging errors hinder reliable correspondence. In dense urban areas, deep learning models frequently merge adjacent buildings—especially within narrow gaps—producing structural inconsistencies that lead to change detection errors. We propose a post-processing method integrating Local Rank-Based Prior Calibration, which reinterprets Softmax probabilities as percentile-based local ranks, with Graph-Cut refinement for structural correction. The refined mask is matched with historical building data to classify four change types. Experiments using aerial imagery from Seoul show that the method reduces structural errors, lowering under-segmentation from 51.64% to 22.02% and improving IoU from 0.748 to 0.759. In change detection, it increases the mean F1-score from 0.522 to 0.608 and improves all classes, including new construction, whose F1-score rises from 0.269 to 0.707. Ablation studies confirm that calibration and graph-based refinement both contribute to the improvements. These results show that stabilizing segmentation outputs enhances the reliability of building-level change detection in dense urban environments.

OpenStreetMap (OSM) is a representative open geospatial platform that provides free access to major spatial objects, including buildings worldwide, constructed through crowdsourcing-based manual digitization. However, subjective differences among contributors and the absence of unified quality control standards have led to the accumulation of positional offsets and boundary shape errors in building polygons. To address this issue, studies using deep learning-based semantic segmentation for OSM quality improvement have been conducted. Nevertheless, Transformer-based segmentation models exhibit an under-segmentation tendency that merges adjacent buildings into a single object, along with limitations in precise boundary delineation. To overcome these challenges, this study proposes a two-stage framework that integrates SegFormer, which excels in global context recognition, with SAM 2, which is capable of precise boundary segmentation. In the first stage, SegFormer semantically segments building regions from a true orthoimage, and in the second stage, SAM 2 infers object-level precise boundaries using the bounding boxes of OSM polygons as box prompts. The two results are combined into a prior probability map, enabling uncertain boundary regions to be re-evaluated in an unsupervised manner. In experiments conducted over the Suseo-dong area in Gangnam-gu, Seoul, the proposed method achieved a BIoU of 70.40%, an improvement of 23.85 percentage points over OSM building data, with consistent performance gains across all evaluation metrics. This framework offers scalability applicable to any region worldwide without additional label construction, provided that high-resolution true orthoimagery and OSM data are available.

Accurate building footprint extraction is critical for applications ranging from population estimation to disaster management. Although optical imagery provides detailed spectral information, it often struggles with shadows, occlusions, and background clutter in dense urban environments. Lidar data, by contrast, offer precise elevation and structural attributes but face challenges such as variable point density and noise. This study integrates multispectral imagery from the U.S. Department of Agriculture (USDA) National Agriculture Imagery Program (NAIP) with lidar-derived feature height and intensity from the U.S. Geological Survey (USGS) 3D Elevation Program (3DEP) to improve footprint extraction using a U-Net–based deep learning model. A six-band input stack (RGB, near-infrared, height, intensity) was developed, normalized, and tiled for training and evaluation against Microsoft Global Building Footprints (GBF). Results from the Houston, TX test site show that the six-band model achieved a precision of 0.86, recall of 0.88, F1 score of 0.87, and Intersection-over-Union (IoU) of 0.76, consistently outperforming four-band baselines by reducing false positives while maintaining sensitivity. Predictions on withheld Houston tiles confirmed strong within-region generalization, yielded a precision of 0.78, recall of 0.81, F1 score of 0.79, and IoU of 0.66. Qualitative analysis further revealed limitations stemming from both training label quality and vegetation–building confusion. These findings demonstrate the complementary value of integrating spectral and structural information for robust building footprint extraction and how domain adaptation strategies can be used to enhance cross-regional transferability.

This contribution presents a benchmarking study of a lightweight deep learning model for automatic pothole detection in asphalt pavements. Accurate and cost-effective identification of surface distresses is essential for road safety and for prioritising maintenance, especially in cities where traditional visual surveys are still predominant. We adapt and train a compact YOLO-based object detection architecture on a dataset of annotated street-level images, covering different lighting conditions, pavement textures and distress severities. The study evaluates how input resolution, confidence thresholds and data augmentation strategies affect detection performance and inference speed, and compares the lightweight model with heavier state-of-the-art detectors. Results indicate that it is possible to obtain competitive accuracy while maintaining real-time processing capabilities on modest hardware, which is crucial for deployment in mobile inspection platforms such as smartphones, dashcams or low-cost onboard units. The paper discusses opportunities and limitations of integrating deep learning into pavement management systems and outlines perspectives for extending the approach to other types of defects and to larger road networks.

Bathymetric Laser Scanning (BLS) enables high-resolution mapping of underwater topography using green-wavelength laser pulses that penetrate the water column. However, precise georeferencing of the BLS data is affected by refraction at the air–water interface, which displaces submerged features and affects conventional strip adjustment methods. This paper introduces an integrated refraction-aware georeferencing workflow that combines refraction correction with trajectory and boresight optimization within a unified adjustment framework. Implemented using the scientific OPALS laser scanning software, the workflow starts with direct georeferencing of uncorrected laser returns, derives a water surface model, applies Snell’s law-based refraction correction, and performs iterative strip adjustment until convergence. The approach was validated using UAV-borne topo-bathymetric LiDAR data from Lake Alm (Almsee) in Upper Austria, captured with a \emph{RIEGL} VQ-840-GE sensor system. Comparative analysis across multiple processing scenarios demonstrates that the proposed integrated method significantly improves internal consistency between overlapping flight strips. The residual height discrepancies, quantified by the median absolute deviation were reduced from 4.5 cm using standard processing workflows to 2.1 cm with the integrated approach — an improvement exceeding 50%. A single processing pass was sufficient for the relatively calm conditions of the test site, though iterative refinement may benefit more dynamic water surfaces. The presented methodology is generic and can be embedded in any laser scanning framework supporting modular georeferencing and refraction correction.

Coastal defense structures, such as breakwaters and groynes are an integral part of coastal engineering. These structures reduce the impact of waves and decrease beach erosion, but due to the constant forces to which they are exposed, repeated monitoring and evaluation is vital to the analysis of their structural integrity. However, coastal defense structures are most often located in the turbulent waters of the surf zone, which characteristics pose severe challenges for current methods. For example, waves pose challenges for image-based analysis, shallow-water limits sonar-based measurements, and currents, represent hazardous environments for surveying personnel. Here, recent advances in topo-bathymetric LiDAR have improved the ability to map data above and below the water surface within the same survey. In the field of structural engineering, point cloud data is already a commonly used information, and thus its applications in the monitoring of coastal defense structures present a natural extension of existing structural monitoring methods. Therefore, this study presents an automatic method for the detection of coastal defense structures with bathymetric LiDAR. The surveyed area consists of multiple groynes located along the Polish coast, which were surveyed using an airplane-based topo-bathymetric LiDAR scanner. The presented method then leverages the echo ratio and repeated clustering to extract the groynes from the data. We evaluate the extracted structures in comparison to manually annotated data. The results of this evaluation display a balanced accuracy of 92%, indicating an overall match with the reference data, but showing challenges and improvements for future work.

A state-of-the-art LiDAR sensor is assessed in terms of the accuracy, described as the sum of trueness and precision, of terrestrial and submerged points. The reference, against which the LiDAR data are evaluated, are conducted with a total station and can be assumed to show an uncertainty of less than 1 cm even for the submerged points. We find that the GNSS-based data set shows a systematic bias of about (-4, 7, 7) cm which can be defined as trueness and does not represent the quality of the LiDAR sensor but mostly of geo-referencing. The precision, which is a measure mostly influenced by the LiDAR sensor itself, is at 0.8 to 2.0 cm for terrestrial points and slightly worse with 1.1 to 2.6 cm for bathymetric points. Our study considers depths of up to 3 m and uses more than 300 points for the assessment.

Worldwide coastal areas comprise environmental triple points (air, land and seawater) that cope with coastal risks at unprecedented rates of change. Wind- and wave-related acute hazards add up to the chronic sea-level rise on interface zones that increasingly host human population and assets. Those societal challenges need to be overcome using the most discriminant and finest remote sensors.

We present an innovative two-step methodology to produce an ultra-high spatial resolution (UHSR) topobathymetry using a fusion of a RGB camera mounted on an aerial drone with a multispectral satellite imagery provided with very high temporal resolution. The fusion relied on a DJI Zenmuse P1 (0,08 m pixel size) borne by a DJI Marice 300 RTK, the PlanetScope SuperDove imagery, provided with eight bands at 3 m, and linear or nonlinear (neural network with two hidden layers endowed with three neurons, each) regression. Once the fusion achieved, both topography and bathymetry were mapped using, either the digital surface model (DSM) derived from the drone-derived photogrammetry, or the DSM combined with the UHSR SuperDove imagery. Both datasets served as predictors to model a digital topobathymetric terrain LiDAR response using linear or neural network regression.

The best drone-satellite fusion was completed by the bandwise neural network regression, ranging from R2test of 0,79 for the purple to 0,94 for the red edge band. The UHSR topobathymetry has been mapped by merging the topography and the bathymetry, distinctly predicted by the combination of the DSM with the UHSR Superdove imagery (R2test of 0,68 and 0,92, respectively).

Rocky coastal environments are ecologically important areas where land and sea processes interact in complex ways. Monitoring these zones is challenging, as they include steep cliffs, partially submerged features, and narrow transition areas where traditional surveying methods often struggle. Several European environmental directives now emphasize the need for regular observation of these coastal systems, increasing demand for practical and accessible surveying tools.

This work presents the development and initial testing of a small uncrewed surface vehicle (USV) designed to collect images above and below the water surface at the same time. The platform is based on a commercially available catamaran-style drone and carries two GoPro cameras mounted on a rigid vertical rod, with one camera positioned above the water and the other just below it. Both cameras are synchronized using GPS time, and the system incorporates a PPK-capable GNSS receiver for improved positioning. The payload is wireless and modular, allowing the platform to be deployed quickly.

The main contribution of the system is its ability to document the air–water boundary in a single pass, reducing issues related to changing meteorological and sea conditions. The paper also discusses how the platform was tested at a rocky site in Sardinia and outlines the types of data that can be obtained for environmental mapping.

The approach aims to offer a low-cost, flexible option for coastal monitoring.

Underwater laser scanners and air-borne laser scanners offer considerable potential for high-resolution monitoring of fine-scale underwater structures in shallow, clear waters. An underwater laser scanner mounted on a vessel is used for kinematic data acquisition in coastal waters. Additionally they are surveyed by an air-borne laser scanner. In this investigation, the resulting point clouds from both systems are analyzed in terms of their performance and achievable relative geometric quality.

The spatial accuracy and resolution of Multibeam Echosounder data are inherently lower than those of high-resolution underwater LiDAR measurements. However, while Multibeam Echosounder provides wide coverage and extensive historical availability, LiDAR is costly and covers relatively small areas. In this study, we propose an innovative approach to enhance Multibeam Echosounder resolution using a Super-Resolution Generative Adversarial Network with direct comparison to LiDAR data for accuracy assessment. The methodology involves converting Multibeam Echosounder data into grayscale format using various depth gradient techniques, analyzing differences in submarine geomorphology through calculations of slope and aspect, and evaluating statistical accuracy. The results show that the Super-Resolution Generative Adversarial Network model successfully improves Multibeam Echosounder resolution, producing data that closely correspond to LiDAR measurements, particularly in flat, sandy seabed areas. In contrast, regions with complex or rocky terrain exhibited more pronounced deviations, especially in aspect metrics, emphasizing the challenges associated with maintaining topographic orientation throughout the resolution enhancement process. The main conclusion is that enhancing Multibeam Echosounder data using Super-Resolution Generative Adversarial Network enables broader utilization of existing datasets to generate high-resolution models, offering a more cost-effective and accurate solution for seafloor mapping in areas where LiDAR data are unavailable.

Solar-induced fluorescence (SIF) has become a promising remote sensing proxy for photosynthetic activity and thus plant health, but its broad application across vegetation types and climate regimes remains underexplored. Here, we present the first continental-scale assessment of seasonal SIF signatures for 33 vegetation types across 24 climate zones in the contiguous United States, enabled by a new open-access visualization tool. The analysis uses TROPOMI satellite SIF data (2019-2021), along with MODIS-derived gross primary productivity (GPP), normalized difference vegetation index (NDVI), and vapor pressure deficit (VPD). Our results show that SIF has consistently stronger and more reliable correlations with GPP than NDVI across vegetation types and environmental conditions. This relationship remains robust even under high VPD conditions (except for several perennial crops), confirming the ability of SIF to track productivity even in dry environments. While NDVI retains structural sensitivity, it often decouples from GPP under stress, particularly in arid climates and perennial crops. We also identify clear differences in SIF-NDVI and GPP-NDVI relationships by vegetation type and climate, with NDVI showing limited responsiveness to dynamic changes in canopy physiology. Despite the coarse spatial resolution of TROPOMI, these results demonstrate the feasibility of constructing climate-specific SIF signatures for agricultural and ecological monitoring. By identifying these climate-specific signatures at the continental scale, this work highlights the value of SIF for climate-smart crop management, productivity assessment, and satellite-based ecosystem modeling.

Accurate estimation of gross primary production (GPP) is fundamental for quantifying the terrestrial carbon cycle. However, coarse-resolution products often fail to capture fine-scale spatial variations in carbon uptake across heterogeneous landscapes. While recent studies have begun to employ 10 m Sentinel-1 and Sentinel-2 imagery, they typically reduce these data to pixel-wise spectral indices, discarding the two-dimensional spatial structure (canopy architecture, land-cover transitions, within-stand heterogeneity) that the imagery encodes. This study investigates whether explicitly exploiting this spatial context via convolutional neural networks yields robust, transferable gains over tabular machine-learning baselines. We curate a quality-controlled dataset of 23,528 eight-day multi-sensor composites from 222 AmeriFlux sites (2015–2025), evaluated under site-wise cross-validation, temporal generalisation, and geographic transfer to an 18-site upper Midwest forest holdout. Under temporal transfer to unseen years (2023–2025), the best convolutional model achieves R² = 0.77 and RMSE = 1.95 gC m⁻² d⁻¹, an 18.6% RMSE reduction over ridge regression (R² = 0.65, RMSE = 2.40 gC m⁻² d⁻¹). Although this advantage narrows under geographic transfer to structurally novel regions (R² = 0.59 vs. 0.54), the convolutional models still outperform all tabular baselines. Spatial structure at 10 m therefore supports more robust temporal generalisation than spectral aggregates alone.

Accurate gross primary productivity (GPP) estimation is crucial for understanding ecosystem function and the global carbon cycle. Remote sensing offers promising GPP proxies, including solar-induced chlorophyll fluorescence (SIF) and the structural proxy NIRvP. However, their performance and underlying drivers of effectiveness vary significantly across biomes. This study comprehensively evaluated the accuracy and limitations of SIF and NIRvP against flux GPP across diverse biomes (CRO, GRA, DBF, ENF), also investigating physiological and structural controls on LUE.

We found that proxy performance was highly biome-specific. Notably, the removal of canopy escape probability (fesc) from observed SIF (SIFobs) to derive total emitted SIF (SIFall) did not consistently improve, and sometimes even diminished, its correlation with GPP, particularly in CRO and GRA. Furthermore, we elucidated distinct dominant controls on seasonal LUE variations: apparent SIF emission yield (ΦF×fesc) was paramount in ENF, while canopy structure (fesc) predominated in CRO, GRA, and DBF. Seasonal analysis in ENF further revealed a temporal decoupling, with fesc decline lagging LUE in winter, and ΦF failing to track autumnal LUE reductions. These findings underscore the biome-specific necessity for optimal GPP proxy selection, establishing a robust scientific foundation for improved remote sensing monitoring.

Accurate quantification of gross primary productivity (GPP) is essential for understanding carbon dynamics under climate change. However, satellite-based GPP estimates face spatial–temporal trade-offs, limiting accuracy in heterogeneous landscapes. To overcome this challenge, we proposed a novel framework named UNified, high-resolution Intelligent carbon QUantification and Explanation (UNIQUE), which produces daily 30 m GPP maps by integrating spatial relationships between 500 m MODIS and 30 m Landsat imagery. UNIQUE consists of two components. First, two AI models were trained using MODIS- and Landsat-based vegetation indices combined with meteorological reanalysis data and validated with 309 eddy-covariance flux tower observations across the Northern Hemisphere. The Light Gradient Boosting Machine (LGBM) showed the best performance, achieving r = 0.80 and RMSE = 2.47 gC/m²/day for MODIS-based GPP, and r = 0.83 with RMSE = 2.43 gC/m²/day for Landsat-based GPP. Second, a diffusion-based deep learning model was used to downscale MODIS-based GPP to 30 m resolution. The diffusion model from MODIS to Landsat GPP exhibited good performance, demonstrating an RMSE of 2.12 gC/m²/day for the testing sites. The proposed approach enabled the analysis of spatiotemporal characteristics of GPP across different plant functional types, facilitating enhanced high-resolution carbon flux monitoring in diverse ecosystems.

Insights into crop productivity have long been of great interest to almond growers, as they enable effective planning to optimise economic returns. Advances in sensor technology have made it possible to collect hyperspectral imagery, which captures detailed information across a continuous range of wavelengths and has become a powerful tool for assessing crop physiological status. Solar-induced chlorophyll fluorescence (SIF), along with other plant pigments and structural traits retrieved through radiative transfer modelling, can effectively track crop photosynthetic activity. However, the ability to quantify SIF is strongly influenced by the spectral resolution of the sensor.

This study examines how the spectral resolution of airborne hyperspectral sensors affects the ability to explain yield variability in a commercial almond orchard, by comparing SIF derived from the 760 nm and 687 nm oxygen absorption bands at different spectral resolutions.

Green roofs (GRs) are constructed systems that replicate natural ecosystems and provide runoff reduction, cooling effect, habitat support and improved air quality services. Over 1,000 GRs have been constructed in Toronto since the GR Bylaw was enacted. As they are dispersed and primarily small-scale stormwater assets on private properties, it is crucial yet difficult to assess these roofs' condition to ensure they continue to deliver the desired advantages and adhere to maintenance regulations. To maintain green stormwater infrastructures in ideal conditions, it is advised that the vegetation be maintained with 80% coverage. GR vegetation experiences plant loss, water stress, and other maintenance concerns requiring regular inspections. This study presents a framework to enable remote assessment of green roof conditions utilizing Sentinel-2A satellite imagery, which captures images every five days. This method overcomes logistical challenges associated with drone imagery inspections, which are limited in frequency and require permits. The study was conducted using Google Earth Engine, focusing on the intra- and inter-annual variation of four GR modules. The study assessed the vegetation health from 2018 to 2025 using NDVI, EVI and NDMI, highlighting the long-term dynamics and distribution of GR vegetation. The results present the effectiveness of NDVI, EVI, and NDMI in assessing plant coverage and moisture content, with low NDMI being an important factor resulting in low NDVI and EVI. The study contributes to the potential of satellite images for scalable and continuous monitoring of GRs and supports efficient and complementary inspection.

Early Wildfire detection using AI remains challenging in environmental monitoring, particularly when the approach should be flexible enough to handle different sensors and different landscapes. This study presents a multi-stage deep learning framework for real-time smoke and fire detection using imagery from fixed tower cameras and UAVs. The proposed system employs YOLOv11 as the primary detection model for high-speed inference, com-plemented by Faster R-CNN for precision benchmarking and cross-architecture evaluation. Together, these models support an in-depth analysis of detection accuracy, robust-ness, and computational efficiency across diverse envi-ronmental conditions.

An end-to-end pipeline has been developed, integrating real-time image acquisition, asynchronous message han-dling through RabbitMQ, and performance logging via InfluxDB, enabling continuous model evaluation under near-operational conditions. Experimental results indicate that while YOLOv11 achieves high frame-rate perfor-mance and strong detection capability, it remains suscepti-ble to false positives in visually ambiguous scenarios such as haze, fog, or low-contrast backgrounds, where contex-tual patterns closely resemble smoke. Faster R-CNN serves as a complimentary reference to quantify localiza-tion accuracy and analyse error propagation, facilitating threshold tuning and model interpretability.

The presented framework bridges the gap between aca-demic model development and field-deployable fire sur-veillance systems. It establishes a reproducible, scalable foundation for real-time decision support in forest watch-tower networks and autonomous UAV missions aimed at early wildfire detection and response

Wildfires induce significant changes in forest structure and the surface reflectance characteristics. This study evaluated the effectiveness of using airborne LiDAR Intensity data to delineate surface fire burn areas in wildfires. We extracted ground returns from both coniferous and deciduous forests and conducted qualitative assessment of Intensity through Intensity images, as well as statistical evaluation using the non-parametric Mann–Whitney U test to compare burned and unburned areas. We compared the median and standard deviation of Intensity at a 10-m mesh scale, calculating standard deviation at a finer 0.5-m mesh resolution. The results revealed significant differences between the two groups. As a result, a significant difference was observed between the two groups. The effect size r for the median in deciduous forests ranged from 0.55 to 0.84, while the effect size r for the standard deviation in coniferous forests ranged from 0.32 to 0.47. Both indicated a medium to large effect. These findings suggest that LiDAR Intensity can effectively identify surface fire burn areas even under heterogeneous forest floor conditions. The proposed method has the potential to contribute to enhancing post-fire monitoring using airborne LiDAR.

This study investigates two decade (2000-2020) of Aboveground Biomass dynamics in the biodiversity hotspot of Western Ghats, India, focusing on the impacts on forest fire and climate variability. Using machine learning approaches with GEDI LiDAR data and MODIS satellite imagery, we developed a robust annual AGB model. These analysis reveals a consistent decline in AGB across Kodaikanal and Nilgiris. Results shows that rising temperature and vapor pressure deficit are the key driver for increase in burn are and fire intensity. These are pushing carbon rich evergreen forests toward a critical transition from carbon sink to source. An integrated Structural Equation Model confirms that the dominant role of climate in driving fire regimes and subsequent biomass loss. This research provides a critical scientific foundation for fire adaptive forest management and carbon accounting in vulnerable tropical ecosystem.

Wildfires are longstanding natural phenomena with significant impacts on ecosystems and communities. In recent years, Canada has experienced particularly severe wildfire effects, especially in British Columbia (BC), which has endured prolonged and impactful wildfire events. However, there is currently no specialized wildfire time-series dataset for BC that considers long-term temporal sequences and multiple driving factors, which are essential for data-driven approaches. To facilitate future research on data-driven wildfire risk and spread prediction, we have developed a dataset covering the entire BC province, encompassing 683 wildfire events from 2002-2023 at 500m resolution with daily observations. For each wildfire event, the dataset includes 20 driving factors, including vegetation status, meteorological factors, human activities, topographical features, and active fire detection. Based on this benchmark and similar datasets from other regions, we compared multiple deep learning models, including CNN-based, Transformer-based, and Mamba-based architectures, to explore the effectiveness of existing deep learning models in wildfire risk prediction. We found that model F1 scores were below 0.6, indicating that this new dataset presents a challenging non-linear modeling scenario that requires more advanced and tailor-designed deep learning models to improve wildfire risk prediction accuracy.

Wildfires are known as one of the most important natural hazards, adversely impacting the ecosystems and human lives. Monitoring and management of wildfires is necessary to minimize their negative effects. Global BA products are widely used to study wildfires, but their accuracy is not constant over different environments. In this study, the MCD64A1 BA product was spatially validated using ground truth maps in a fire-prone Zagros Forest over 2021-2023. Our results indicated that its performance varies temporally, as the Kappa coefficient ranged from 0.04 to 0.69. Overall accuracy was higher than 0.96 percent in all years, indicating that MCD64A1 can be considered as a source for studying wildfires; however, its underestimation should be considered. In the next step, the trend of fire and its relationship with precipitation (i.e., obtained from the CHRIPS dataset) were analyzed in three forest ecosystems from 2001 to 2024. In two regions, Marivan and Kermanshah, wildfires experienced an increasing trend, in contrast to the other region, Shiraz, where they decreased over time. Analyzing the correlation between fire and precipitation revealed that spring precipitation is more connected to BA than annual precipitation. Comparing the results of the three regions showed that this matter is also region-related, and the results of one region cannot be referred to another. This study provided information on the performance of MCD64A1 in semiarid forests and the wildfire conditions in the Zagros Mountains to aid wildfire management.

This contribution introduces GT-LOD3, a new benchmark dataset designed to advance semantic Level of Detail 3 (LOD3) building reconstruction from UAS-based photogrammetric point clouds. Existing benchmarks primarily focus on mesh- or point-level semantic labelling, façade segmentation, or LOD2-level modelling, but high-quality, geometry-accurate LOD3 ground truth paired with real-world photogrammetric observations are still limited. GT-LOD3 fills this gap by offering paired UAS point clouds and manually modeled LOD3 reference data in CityGML format, enabling research on window-level facade reasoning, geometric regularization, and instance-level shape recovery.

The benchmark currently consists of two subsets featuring different architectural styles and environmental conditions: (1) a urban block in Gold Coast (Lakewood, Ohio, USA), and (2) the Technical University of Munich (TUM) campus. The accompanying LOD3 reference models contain explicit window geometry, enabling detailed evaluation of both detection performance and polygon-level geometric accuracy.

We further provide a baseline reconstruction pipeline that combines point-cloud semantic segmentation, facade-aligned 2D projection, window region extraction, and geometric back-projection into CityGML. An evaluation protocol is presented including pixel-level metrics (IoU, precision, recall, F1) and instance-level detection metrics based on optimal assignment via the Hungarian algorithm.

Building wireframe reconstruction from LiDAR faces challenges due to sparse and incomplete point cloud data. We present LoD2-Former, a multi-modal Transformer architecture that fuses aerial LiDAR and optical imagery for end-to-end 3D roof wireframe reconstruction. Unlike existing point-cloud-only methods, our dual-backbone approach with bidirectional cross-modal attention leverages complementary geometric and visual information. Experiments on two datasets show consistent improvements in edge detection metrics, with edge F1-scores increasing from 0.874 to 0.899 on Tallinn and 0.968 to 0.974 on Roof-Intuitive, while substantially boosting corner recall (0.630 to 0.729) in complete-data settings. We also contribute a curated multi-modal subset of Building3D with aligned LiDAR and aerial imagery to facilitate future research.

In this paper, a method exploiting aerial LiDAR point clouds to build realistic building meshes suitable for electromagnetic simulation is proposed. One of the main challenges lies in reconstructing regularized building meshes with low polygonal density. Optimization-based methods, commonly used for building reconstruction from point clouds, are highly data-driven, making the quality of results dependent on the quality of input data. Aerial LiDAR scans can be incomplete or sparse, for instance due to occlusion. A novel LoD2 buildings reconstruction method based on deep learning is proposed, assuming that deep learning methods are more robust to incomplete or sparse data than optimization-based methods. A parametric building model is introduced, based on the Weighted Straight Skeleton algorithm, which generates realistic roofs from a building footprint and an associated set of slopes, and subsequently extrudes the roof to the specified building height. This parametric approach guarantees that a given set of parameters (height, footprint and slopes) produces a regularized building mesh with low polygonal density. A multimodal model, named Point2WSS, was trained to recover the variable number of building's continuous parameters from its corresponding point cloud. This approach enables the generation of realistic building meshes suitable for electromagnetic simulation, if the predicted parameters accurately approximate real-world values.

The modeling of buildings suffers from a dichotomy between generic and specific representations: the lack of domain knowledge in flexible models that can represent many shapes, and the restricted geometry of pre-specified parametric building primitives. To fill this gap, we propose using general boundary representations enriched with automatically recognized and enforced geometric constraints derived from human-made regularities. The proposed reasoning process relies on the statistics of the planar point groups extracted from airborne-captured point clouds. Hence, a chosen significance level is the only process parameter. To enforce the creation of sound solids, we apply manifold constraints for the generation of the boundary representations. The feasibility and usability of the approach are demonstrated by evaluating an airborne-captured laser scan containing approximately 7,600 buildings over an area of 50 km^2 featuring both inner-city and rural landscapes.

We present P3, a large-scale multimodal dataset for building vectorization, including aerial LiDAR point clouds, aerial images, and vectorized 2D building outlines, collected across three continents. P3 contains over 10 billion LiDAR points with decimeter-level accuracy and RGB images at a ground sampling distance of 25 centimeters. While many existing datasets focus on the image modality, P3 offers a complementary perspective by incorporating dense 3D information. We demonstrate that LiDAR point clouds serve as a robust modality for predicting building polygons, both in hybrid and end-to-end learning frameworks. Moreover, fusing LiDAR and imagery further improves accuracy and geometric quality of predicted polygons. The P3 dataset is publicly available, along with code and pretrained weights of three state-of-the-art models for building polygon prediction at https://github.com/raphaelsulzer/PixelsPointsPolygons .

Accurate building height estimation plays a crucial role in large-scale 3D urban reconstruction. However, conventional stereo matching approaches often suffer from mismatches around building edges, leading to unreliable height retrieval in dense urban areas. To address this issue, this paper presents a novel method for building height estimation based on contour vector registration integrated with the vertical line locus technique. The proposed framework first automatically matches building contour vectors extracted from stereo high-resolution satellite images. Then, for each paired contour, a range of candidate heights is searched using a rational function model to project the reference contour from the image space to object space and then reproject it onto the conjugate image. The elevation that maximizes the overlap ratio between projected and paired contours is identified as the optimal roof elevation. Building height is subsequently derived by subtracting the ground elevation from the estimated roof elevation. Experiments conducted on SuperView-1 (SV-1) satellite stereo images over Jiuyuan District, Baotou, Inner Mongolia, China, demonstrate the effectiveness of the proposed method. The resulting building height estimates achieve a root mean square error of 0.84 m compared to manual measurements, showing strong agreement (r = 0.9993). The proposed contour-based stereo registration approach provides a robust and efficient solution for building height extraction from high-resolution satellite data, supporting precise urban 3D modeling and large-scale spatial analysis.

In 2024, the Singapore Land Authority (SLA) commissioned Woolpert to conduct a large-scale aerial mapping initiative under the National 3D Mapping Programme to support Smart Nation applications, urban planning, and geospatial analytics. This project, executed between 2024 and 2025, delivered high-resolution imagery and LiDAR datasets across approximately 750 km², covering mainland Singapore and offshore islands. This was the third epoch of 3D mapping in Singapore with previous surveys conducted by Woolpert (then AAM) in 2014 and 2019

Accurate, high-resolution land use and land cover data are critical for effective environmental monitoring, watershed management, and sustainable urban and peri-urban planning within rapidly urbanizing regions such as the Toronto and Region Conservation Authority (TRCA) jurisdiction in Ontario, Canada. TRCA has conventionally relied on manual mapping approaches to delineate its LULC inventory; however, this method is labour-intensive and prone to temporal inconsistencies across updates. To address these challenges, we developed TRCA-AutoMap, a deep learning-based automated mapping framework to generate fine-scale LULC products using 3-m PlanetScope imagery. TRCA-AutoMap integrates two principal modules. The first module is designed to enhance the model’s ability to detect and differentiate objects across spatial scales. By leveraging multi-extent feature encoding and pyramid pooling, the convolutional neural networks capture both fine-texture and contextual information, thereby improving segmentation accuracy and spatial coherence . The second module focuses on optimizing the model’s understanding of varying imaging conditions. It utilizes a group of autoencoders to mitigate radiometric and environmental differences among input images, thereby maintaining the model's reliability across varied lighting conditions, sensor types, and atmospheric conditions. This process enhances the stability of PlanetScope imagery over time and consistency between different scenes. The framework significantly reduces manual processing effort, ensures classification consistency, and supports annual LULC updates. Quantitative and visual evaluations confirm that the model accurately captures fine-scale vegetation heterogeneity and urban expansion dynamics.

With the advancement of urbanization and digital twin city development, urban 3D data are characterized by large volume, heterogeneity, and structural complexity. Traditional spatial data management methods face limitations in hierarchical organization, retrieval efficiency, and redundancy control, and the lack of a unified spatial coding system hinders multi-source data integration. This paper proposes a method for urban 3D data management and representation based on BeiDou grid coding and adaptive voxel modeling. The method converts point cloud data from local coordinates to the 2000 National Geodetic Coordinate System, applies 36-bit 3D BeiDou grid coding, performs adaptive octree voxel partitioning based on point cloud density, elevation variation, and class entropy, and binds spatial, geometric, and semantic attributes at the voxel level. Using the SensatUrban dataset, the method is compared with fixed-resolution voxel modeling, latitude-longitude indexing, and R-tree indexing in terms of voxel quantity, data storage, and retrieval time. Results show that it reduces voxel count by 28.1% and storage volume by 13.6% while maintaining high-precision representation, and the BeiDou grid-based indexing significantly improves query efficiency and stability. The proposed approach balances visualization quality and computational efficiency, providing an effective solution for large-scale urban 3D data management.

Smartphone 3D-scanning apps are becoming popular tools for heritage documentation, but their accuracy and reliability remain unclear. This contribution presents controlled laboratory experiments using several iPhone-based scanning applications, comparing their point clouds to high-precision reference data. The study evaluates geometric accuracy, completeness, and reconstruction geometric stability, highlighting the strengths and limitations of current mobile scanning solutions. Practical recommendations are provided for heritage professionals and field teams, along with future directions for improving smartphone-based documentation using AI-enhanced depth estimation.

The generation of 3D urban morphology models from 2D urban morphology maps has been widely explored. Traditional methods use modelling software, such as Rhino, which lack georeferencing, elevation, and automation. In this study, we developed an open-source Python framework for automatic generation of 3D city blocks, including elevation, from 2D colour-graded building heightmaps and urban morphology input. We utilised the UT-GLOBUS and GlobalBuildingAtlas building datasets to generate heightmaps and retrieved other urban morphology features, such as waterbodies, parks, roads, and trees, from OpenStreetMap to form the input raster patches. The framework generates height and colour maps based on the input features, which are extruded in 3D and exported into multiple standard 3D GIS formats such as CityGML and CityJSON. Six global cities: Sydney, New York, London, Rio de Janeiro, Hong Kong, and Singapore, were modelled to demonstrate the framework’s applicability. Validation includes qualitative comparison with Google Earth 3D data and quantitative comparison against official LiDAR-derived DSMs for four cities. Quantitative results show moderate height errors and good spatial agreement of building footprints, reflecting the expected differences between simplified LoD1 block models and detailed DSM representations. Our framework results show promising potential in the field of 2D to 3D mapping for the creation of 3D city models for urban climate modelling and environmental analysis. The generated 3D models can be downloaded at https://doi.org/10.5281/zenodo.17620303.

Among the image-based methods, traditional photogrammetry is a consolidated 3D reconstruction technique able to provide highly accurate metric products, widely exploited in many domains for documentation and mapping purposes. The reconstruction capability of this technique is, however, conditioned by the characteristics of the captured scene, with high performance in well-textured areas and limits when non-collaborative surfaces, such as reflective or transparent objects, are present. In such cases, the photogrammetric reconstruction is often affected by noise, incomplete geometry and artifacts, reducing its final reconstruction quality. In recent years, different AI-based reconstruction methods have emerged as alternative (or complementary) 3D reconstruction and rendering solutions. In particular, 3D Gaussian Splatting (GS) has demonstrated impressive capabilities in rendering photorealistic scenes in challenging situations with high visual fidelity. However, its application in large-scale scenarios or when highly accurate 3D metric products are required is still limited, due to the hight computational resources needed and the intrinsic optimization of GS methods for photometric rendering quality. To address these bottlenecks, this work proposes a hybrid reconstruction pipeline, leveraging the strengths and benefits of each technique. The method exploits the accurate geometry of photogrammetry in well-textured regions and the higher GS capabilities to improve completeness and visual aspect in areas featuring non-collaborative surfaces. A fusion strategy is proposed to combine the two products into a single 3D model, presenting results on two aerial and one terrestrial dataset.

Neural Radiance Fields (NeRFs) (Mildenhall et al., 2020) have revolutionized novel view synthesis, but standard formulations assume straight rays in a single, homogeneous medium. In underwater scenarios, refraction at the air–water interface leads to bent light paths and, if ignored, to distorted 3D structure and missing underwater points. Refraction-aware NeRF variants such as NeRFrac (Xue et al.,2022) demonstrate the benefit of modeling refraction, but are limited to a single underwater medium and standalone implementations. Recent work has applied NeRFrac to through-water reconstruction (Brezovsky et al., 2025), introduced a simulation framework for two-media scenes (Schulte et al., 2025). Building on these ideas, we introduce the general concept of a twomedia NeRF and demonstrate its integration into the Nerfstudio framework (Tancik et al., 2023) with the goal of extracting metrically meaningful underwater point clouds rather than only improving image-based metrics.

Many geospatial workflows critically depend on high-fidelity 3D point clouds for applications such as change detection, orthophoto generation, and modeling. However, NeRF-generated point clouds often suffer from sampling inefficiencies inherent in the predominant random pixel sampling approach. We identify spatial redundancy as one such inefficiency: random sampling has the inevitable consequence of sampling large, low-texture patches more frequently than detailed, high-frequency textured regions. As a result, low-texture areas turn to be oversampled and other pixels remain unsampled -- regardless of their importance to the reconstruction task.

To overcome this, we propose CENS (Coverage-Efficient Non-Redundant Sampling), a deterministic pixel sampling strategy that maximizes spatial coverage, eliminates intra-image sample repetition, and ensures reproducibility via structured initialization.

Evaluated on the Jamtal valley dataset, CENS achieves comparable geometric accuracy (C2M: mean = -0.0027 vs. -0.0011 m; standard deviation = 0.027 vs. 0.028 m) using 50% fewer training steps (11,232 vs. 22,464), while yielding 28.2% more points, higher orthophoto fidelity, and improved point cloud completeness. Beyond CENS, we also explored NeRFs for ALS point cloud simulation, achieving realistic occlusion patterns and accuracy within UAV photogrammetry standards (Vertical RMSE} = 24 mm; Horizontal RMSE = 17 mm).

Crucially, CENS positions NeRFs as a scalable, practical solution for geospatial point cloud and orthophoto generation, advancing them toward real-world mapping workflows, and integrates seamlessly into NeRFStudio.

This research presents an innovative multi-modal framework for the explicit 3D reconstruction of building thermal environments to diagnose facade defects. We propose a framework which is centered on a dual-branch Neural Radiance Field (NeRF) architecture, which effectively fuses fine-grained geometric information from RGB data with precise quantitative thermal data from TIR data. For practical diagnostics, the framework integrates the Signed Distance Function (SDF) to implicitly learn a high-fidelity surface representation. Subsequently, a final, explicit triangular mesh is extracted from this implicit field using the Marching Cubes algorithm. The resulting model achieves geometric accuracy and thermal fidelity, enabling the clear visualization, localization, and analysis of thermal anomalies such as thermal bridges, cavities, and moisture ingress in their correct spatial context.

3D vision foundation models, which directly regress 3D geometry from 2D images in an end-to-end manner, have recently attracted growing attention in the computer vision community. However, their potential for oblique 3D reconstruction has not been systematically evaluated. To this end, we establish an automated evaluation pipeline to benchmark these models on oblique imagery. Our experiments reveal that: benefiting from the powerful zero-shot generalization, 3D vision foundation models can robustly estimate camera parameters and generate dense point clouds under sparse-view and low-overlap conditions, with some rivaling traditional photogrammetry configured with redundant observations. Counterintuitively, two-view reasoning foundation models employing explicit PnP-RANSAC for global alignment consistently outperform multi-view reasoning foundation models inferring multi-view relationships via implicit attention mechanism when processing more than 2 views. Notably, incorporating known camera parameters as conditioning inputs, which act as weak supervision rather than rigid geometric constraints, yields only marginal accuracy improvements. Based on ViT architecture, these foundation models face scalability bottlenecks to large-scale and high-resolution oblique imagery, and their prevalent ideal pinhole camera assumption still makes explicit distortion correction an unavoidable preprocessing step.

Three-dimensional (3D) reconstruction from satellite imagery is a critical research topic in the fields of remote sensing and geoinformation science. Although 3D Vision Foundation Models (3D VFMs) have demonstrated remarkable performance in reconstructing natural scenes, their capability to handle high-resolution satellite imagery has not been systematically evaluated. This study presents a comprehensive assessment of seven representative 3D VFMs for satellite-based 3D reconstruction and integrates four point-cloud alignment strategies. Rigorous comparisons were conducted against high-precision LiDAR-derived Digital Surface Models (DSMs) using two publicly available multi-view satellite datasets--WHU-TLC and MVS3D. Experimental results show that, on the high-resolution MVS3D dataset, the Depth Anything v2 (DAV2) model combined with the Affine alignment strategy achieved the best overall performance, producing DSMs with a Mean Absolute Error (MAE) of 1.75 m and a Root Mean Square Error (RMSE) of 3.24 m, corresponding to accuracy improvements of 8.4 % and 13.6 %, respectively--significantly outperforming all other model-strategy combinations. In contrast, on the lower-resolution WHU-TLC dataset, all 3D VFMs exhibited notable performance degradation, and the reconstructed results showed limited practical value, revealing persistent generalization challenges for current models in low-resolution scenarios. Overall, this study systematically quantifies the performance of 3D VFMs in satellite image-based 3D reconstruction, confirming their strong potential for high-resolution satellite applications and providing valuable insights for enhancing model robustness and generalization across complex urban and low-resolution environments.

High-resolution Earth observation data are crucial for applications such as agriculture, urban planning, and environmental monitoring. Although commercial satellites provide sub-meter imagery, open-access alternatives like Sentinel-2 are limited to resolutions around 10~m ground sampling distance, which is insufficient for many tasks. In this work, we investigate image super-resolution as a method to bridge this gap, enhancing downstream performance on freely available satellite data. We leverage two 16-bit single-band datasets, consisting of Sentinel-2 (20m --> 10m) and Venus (10m --> 5m) images, to train and benchmark state-of-the-art SR methods, including transformer- and diffusion-based approaches, across multiple dataset mixes. These models are evaluated quantitatively using reference-based metrics (PSNR, SSIM) using ground-truth and no-reference scores (FID, NIQE) for native upscaling from 20m --> 10m and 10m --> 5m. We observe that different SR architectures present trade-offs between standard quantitative metrics and perceptual image quality. We further assess their impact on a practical downstream task: field boundary detection from Sentinel-2 imagery. Our experiments demonstrate that SR pre-processing improves quantitative fidelity and downstream task performance, enabling low-resolution satellites to compete more effectively with commercial imagery

Current diffusion models struggle to achieve fine-grained remote sensing imagery (RSI) generation. This limitation fundamentally stems from their reliance on "flattened" text prompts, which overlook the inherent hierarchical structure of RSI. This paper proposes a fine-grained RSI generation method driven by expert knowledge and hierarchical captions. We first deconstruct RSI into a hierarchical "element-relation-scene" caption and employ an automatic caption optimization mechanism, grounded in spatial relation knowledge, to ensure high fidelity. Critically, we introduce a novel hierarchical caption encoding mechanism that efficiently injects decoupled hierarchical caption segments into the U-Net's cross-attention layers. This design enables the model to exert hierarchical and decoupled attentional control over the global scene, spatial layout, and geographical element details during the denoising process. Experiments demonstrate that, when combined with efficient fine-tuning algorithms such as LoRA, our method significantly outperforms traditional single-level captions across all six evaluation metrics, exemplified by the FID metric decreasing from 228.43 to 205.59 and the GSHPS metric increasing from 0.86 to 0.92. This research provides a new paradigm for controllable remote sensing scene generation, establishing an effective link between hierarchical semantic understanding and the progressive generation process of diffusion models.

Semantic segmentation of remote sensing images has attracted considerable attentions. For cross domain semantic segmentation, the images captured at different times inevitably exhibit significant domain and feature gaps. Besides, the labels are precious, given that acquiring adequate annotations is time-consuming and laborious. There are numerous methods to cope with these problems, for example, semi-supervised, weakly supervised learning help for the lack of label, while style transfer and domain adaptation are effective for domain gaps. However, the outcomes are still not ideal. Nearly all methods ignore the combination of image-level alignment and feature-level alignment, while few methods consider class-wise constraint to boost the performance. Towards this end, IFSDA, an image-level and feature-level semantic-aware architecture for cross domain semantic segmentation is put forward. In order to acquire sound outcomes, two branches of alignment strategies are realized by self-supervised learning and generative adversarial learning. Besides, a novel semantic discriminator is utilized in image translation process to optimize class-related information, thereby helping to eliminate the intra-class domain gaps between bi-temporal images and optimize the segmentation results effectively. Experiments on ISPRS 2D Semantic Labeling Contest Dataset have shown the superiority of proposed method over other models.

Reliable estimates of seabird colony size are essential for monitoring population dynamics, yet accurate counts are difficult for expansive colonies on remote Arctic cliffs. Northern fulmars (Fulmarus glacialis) breed in large, unevenly distributed aggregations across extensive, towering cliffs in the Canadian Arctic, posing numerous survey challenges. Side-looking helicopter photo surveys generate thousands of photos where birds are small, variably angled, and of inconsistent sharpness against large, complex backgrounds. We used deep learning to automate fulmar counts in this imagery. Our objectives were to (1) develop an object detection model trained on manually annotated imagery sampled from three Arctic colonies, (2) evaluate model performance, and (3) estimate total size of an entire colony. We trained a YOLOX-based model on >16,000 annotated birds, following a two-stage training approach for small objects interspersed across expansive and heterogenous backgrounds. Compared to ~20,000 additional manual annotations in a sample of the Cape Liddon colony in the territory of Nunavut, the model detected 90% of birds with a 9% false-positive rate (i.e. 90% recall, 91% precision). The model's detection sensitivity was calibrated to achieve a ~1:1 ratio between total model detections (true positives + false positives) and the 'true' count, which required manual annotation of ~15% of the colony imagery. Overall, the model detected 38,723 fulmars across the entire colony, providing a robust estimate of its full population. These results highlight deep learning’s potential to greatly streamline and scale up seabird monitoring in remote polar environments where conventional surveys are constrained.

Air pollutants such as nitrogen dioxide (NO₂) have detrimental effects on human health and ecosystems. It is therefore very crucial to pinpoint the location of high pollutant concentrations over large areas. Ground-based stations, while offering continuous temporal measurements, cannot provide broader spatial coverage for regions like cities. This study uses Tropospheric Emissions: Monitoring Pollution (TEMPO) satellite observations and a machine learning model to estimate high-resolution surface-level NO₂ concentrations over the Greater Toronto Area (GTA), Ontario, Canada. The random forest regression model was trained with input parameters such as hourly tropospheric NO₂ vertical column density (VCD) values and boundary layer height (BLH), which are the two most effective parameters in feature importance. The model achieved a coefficient of determination (R²) of 0.84, a root mean square error (RMSE) of 1.703 µg/m³, and a mean absolute error (MAE) of 0.939 µg/m³, indicating strong and reliable predictive performance. The findings of this research can support air quality forecasting, public health studies, and urban planning decisions, especially in regions with scarce ground-based pollutant data.

Automatic updating of topographic maps remains a significant challenge, as current workflows still rely heavily on manual interpretation of airborne data. This study proposes a method for identifying topographic changes by learning object representations from existing maps and using them as reference data for change detection. Map-derived labels are used to train independent 2D and 3D segmentation networks that generate semantic predictions from orthoimages and point clouds. Unlike conventional change-detection approaches that require temporally aligned datasets of the same modality, the proposed method directly compares newly acquired airborne data with existing map vectors. Semantic predictions from both modalities are vectorized and selectively fused into polygon geometries, which are subsequently compared with reference map vectors to identify object-level "from–to" changes. The workflow highlights potential change regions and their predicted semantic classes, allowing operators to focus inspection on relevant areas rather than the entire dataset. Detected changes include both real-world developments, such as new construction and demolitions, and inconsistencies in the reference map caused by outdated or inaccurate delineations. To assess the effect of multimodal integration, the workflow is compared with a 2D-only baseline. The results indicate that integrating 3D geometric information can reduce noisy detections and improve the spatial consistency of candidate change objects, particularly for water and bridge classes.

Point clouds have become essential data for describing real-world objects. Accurate and efficient 3D semantic segmentation plays a crucial role in environment understanding and scene reconstruction. However, current segmentation methods still face challenges from unordered data, high computational complexity, limited scene perception, and insufficient generalization. To address these issues, we propose a local-to-global semantic segmentation method based on a state-space model (LGSSM). Specifically, the proposed method uses three-dimensional serialization encoding to serialize point clouds along the x, y, and z directions, effectively addressing the inherent disorder of point clouds and enhancing spatial representation. Then, the local state space model extracts fine-grained local geometric structural information and the global state space model captures the overall scene representation, improving the modeling ability for both short and long distances. Finally, the serialized context aggregation module is utilized to fuse contextual features to promote spatial semantic consistency. Extensive experiments conducted on ScanNet, ScanNet200, and S3DIS demonstrate that our model achieves state-of-the-art segmentation accuracy compared with other existing methods.

In this paper, we present a novel approach to semantic segmentation of airborne LiDAR point clouds that integrates a hierarchical Gaussian Mixture Model (hGMM) within the Superpoint Transformer (SPT) framework. The hGMM constructs a coarse-to-fine representation of the scene by recursively fitting Gaussian components to spatially coherent subsets of the point cloud, resulting in a hierarchical and structured decomposition that serves as a structured token set for the segmentation objective. While Gaussian Mixture Models (GMMs) can virtually fit any distribution, we constrain their use to structured suburban scenes, where their parametric form is naturally suited to represent planar and ellipsoidal geometries, hence allowing parsimonious mixtures. Experimental results on the DALES benchmark demonstrate that our method achieves competitive performance with respect to state-of-the-art approaches, with notable improvements on classes such as ground and buildings. Results on indoor S3DIS confirm the method's intended specificity to outdoor environments. These findings validate hGMM as a principled and effective alternative to heuristic partitioning techniques, integrating stochastic modelling with transformer-based semantic reasoning in large-scale 3D environments.

Point cloud semantic segmentation through multi-modal fusion provides a fundamental basis for surface observation and visual perception tasks. LiDAR provides precise geometric structural information, while optical images offer rich semantic and textural details. However, existing fusion methods still suffer from limited cross-modal perception and insufficient information complementarity. To address these challenges, we propose a multi-stage LiDAR-image collaborative perception fusion network (MCPFNet) for point cloud semantic segmentation in urban scenes. In the mid-stage, the network introduces a geometric-aware fusion (GAFM) and a semantic-aware fusion module (SAFM) to achieve bi-directional injection of structural and semantic features between LiDAR and image modalities. In the later stage, an adaptive feature fusion module (AFFM) is designed to refine semantic representations through gated weighting and bi-directional attention mechanisms. Extensive experiments demonstrated that MCPFNet achieved the best mIoU scores of 74.51%, 72.10%, and 95.15% on the ISPRS Vaihingen, FRACTAL, and N3C datasets, respectively, validating its superior performance in multi-modal semantic segmentation.

This contribution investigates the cross-sensor and spatial generalization of deep learning methods for 3D semantic segmentation in railway environments. Although current models achieve high accuracy on large benchmark datasets, their robustness under real-world acquisition variability remains insufficiently understood. To address this gap, three state-of-the-art architectures—Point Transformer v3, Swin3D, and MinkUNet—were trained on the SemanticRail3D dataset and evaluated on a newly acquired 120-m railway section captured with three heterogeneous LiDAR systems: a Faro Focus S150+ terrestrial laser scanner, a CHCNAV RS10 handheld device, and a GeoSLAM ZEB Go SLAM-based scanner.

The case-study point clouds were carefully registered, normalized, voxelized, and manually annotated to provide consistent ground truth across sensors. A standardized preprocessing and test-time augmentation pipeline was applied to ensure compatibility with the training domain. Generalization performance was analysed through per-class IoU, cross-model agreement, and sensor-dependent degradation patterns. Results show significant variability across acquisition platforms, with denser, low-noise scans enabling better transferability, while sparser SLAM-based point clouds remain challenging for thin or small components such as overhead wires.

To mitigate cross-sensor variability, an IoU-weighted ensemble strategy was introduced, leveraging complementary model strengths without requiring retraining. This ensemble consistently improved or matched the performance of individual models on the case-study datasets.

Overall, the study demonstrates the importance of evaluating semantic segmentation models under realistic multi-sensor conditions and provides a practical benchmark and methodology for assessing domain-shift effects in railway point clouds.

Accurate semantic segmentation of urban point clouds is fundamental for autonomous driving and city mapping. Recent advances in neural scene representations, such as Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS), have significantly improved photorealistic reconstruction quality. However, 3DGS is primarily designed for small-scale, object-centric scenes with dense viewpoints, and its optimization becomes sub-optimal in large-scale street scenes with trajectory-constrained observations, leading to semantic errors and distorted geometry.

In this work, we revisit NeRF-based scene representation in the era of 3DGS to address these challenges. Our method integrates the explicit and efficient modeling strategy of 3DGS with the surface-constrained sampling nature of NeRF. Specifically, we employ Deformable Neural Mesh Primitives (DNMPs) to jointly encode geometry and semantics, enabling efficient ray–mesh intersection sampling and neural field interpolation. This formulation achieves 3D-annotation-free point cloud semantic segmentation by leveraging rendered image supervision.

Experiments on the KITTI-360 dataset demonstrate that our approach surpasses the Street Gaussians baseline in overall mIoU and across most semantic categories. The improvement mainly stems from reducing semantic errors caused by limited viewpoints during 3D Gaussian optimization, providing a robust and scalable solution for street scene semantic understanding.

This paper presents a fusion framework for the automatic extraction of pole-like road objects—such as traffic lights, road signs, streetlights, and utility poles—from Mobile Mapping System (MMS) point clouds. The proposed method integrates KPConv-based semantic segmentation with geometric–topological reasoning to achieve structural completion and false-positive suppression without retraining or additional annotated data. The framework was trained on 8 km of manually labeled MMS data from the Kinki region, Japan, and evaluated on large-scale unseen data from Hokkaido (≈ 26 km, 2.53 billion points) and the Paris–Lille-3D benchmark (France) acquired with a different LiDAR sensor. The proposed approach significantly outperformed the KPConv baseline. On the Hokkaido dataset, the F₁-score improved from 0.8263 to 0.8689 (+0.0426), successfully reconstructing lamp tops, signal arms, and previously unseen snow delineator posts (snow poles). On the Paris–Lille-3D benchmark, recall increased by 15.5 points, yielding an overall F₁-score gain of +0.0802. The 26 km Hokkaido dataset was processed in less than 13 hours on a single NVIDIA Quadro RTX 8000. These results demonstrate that the proposed deep learning–geometry–topology fusion achieves robust, scalable, and efficient performance across diverse geographic and sensor domains, supporting nationwide road-asset mapping and digital-twin generation.

This study presents a deep learning framework integrating Sentinel‑2, Sentinel‑1, and GEDI LiDAR to map Ethiopia’s canopy height at 10‑m resolution from 2019–2024. A shift‑aware loss function was employed to correct geolocation errors inherent in GEDI L2A footprints, and height‑weighted penalties addressed systematic underestimation in tall forests. Results show a national net gain of 23,537 km² in tree cover >8 m, reversing long‑standing deforestation trends. Gains concentrated in low‑to‑mid canopy strata (<20 m), strongly associated with major restoration interventions including the Green Legacy Initiative (GLI), REDD+, and the Sustainable Land Management Program (SLMP). Losses persist in western and southeastern highlands, driven by agricultural expansion, wildfires, infrastructure development, and large‑scale agricultural investments. This work demonstrates the operational value of multi‑sensor deep learning for near‑real‑time monitoring of restoration outcomes in data‑scarce regions.

Accurate tree-level analysis in forests via LiDAR scanning is essential for biomass estimation, canopy structure assessment, and carbon monitoring, yet remains constrained by the scarcity of large-scale annotated LiDAR datasets and the high cost of manual annotation. To address this, we present a novel approach that integrates 3D tree models with UAV-borne LiDAR simulation to generate synthetic forest point clouds with comprehensive annotations. Our approach generates diverse woodland, open, and closed forest structures, producing Synthetic Forest, a benchmark datasets of three 1 ha scenes containing 38–47 million points each, with densities of 3300–3860 points/m² and average spacing of 2 cm. Each scene contains between 70 and 216 individual trees, along with understory vegetation, deadwood, stumps, rocks, and bushes, all automatically annotated with semantic classification IDs, instance IDs, and tree IDs for volume estimation. The proposed pipeline provides automated, error-free ground truth for leaf-wood classification, instance segmentation, and wood volume estimation. We provide a guideline for generating forest plots and utilizing the datasets for diverse forestry tasks. By eliminating the need for costly field data collection, our pipeline offers scalable, customizable synthetic datasets that accelerate forest inventory. The Synthetic Forest dataset is publicly released via Zenodo (DOI: 10.5281/zenodo.17568131), enabling reproducible research and supporting further developments in forest monitoring and management.

Accurate mapping of tree species is essential for forest monitoring, biodiversity assessment, and ecological applications. Very high-resolution UAV imagery provides detailed structural and spectral information, but species-level segmentation remains challenging due to limited annotated data, complex crown geometries, and strong visual similarity among taxa. Recent Remote Sensing Foundation Models (RSFMs) offer new possibilities by providing transferable representations learned from large, multimodal geospatial datasets.

This contribution introduces a two-phase framework that combines foundation model initialization with multi-temporal UAV imagery to enhance fine-grained forest segmentation. In Phase 1, a DeepLabv3+ network is initialized using FoMo-Net, a ViT-based RSFM pre-trained on the multi-scale FoMo-Bench benchmark. This initialization enables strong generalization from heterogeneous global forest datasets to very high-resolution UAV scenes. In Phase 2, phenological information is integrated by fusing May and September UAV acquisitions through temporal difference composites and pseudo-label refinement, allowing the model to resolve species-specific seasonal patterns.

Experiments on the Québec Trees Dataset, covering 14 species at 0.02 m GSD, demonstrate substantial performance gains. Foundation model initialization improves overall accuracy from 52.79% to 71.21%, while incorporating multi-temporal cues further increases accuracy to 78.21%. The results highlight the complementary roles of structural priors learned by RSFMs and phenological information captured by UAV time series for detailed forest species mapping.

Accurate mapping of forest types and vegetation characteristics is essential for monitoring biodiversity and forest dynamics. Traditional Deep Learning (DL) models trained on Sentinel-2 time series achieve high performance, but require extensive preprocessing and sensor-related fine-tuning. In this study, we evaluate the recently introduced AlphaEarth Foundations (AEF) embeddings, which is global, multi-modal feature representation of the earths surface, for forest mapping in Italy. We compare a) a Multi-Layer Perceptron trained on AEF, b) a Time-Series Transformer trained on Sentinel-2 annual time series and CHELSA climate data, and c) a Cross-Attention fusion model combining both feature sets. Using 5-fold cross-validation in a regression and a classifaction task on two datasets (evergreen broad-leaved tree cover ETC, forest vegetation type FVT) we find that the combined model consistently outperforms the single-source approaches (RMSE = 0.161, Accuracy = 0.757). AEF-based models achieve comparable accuracy to the Sentinel-2-based model while reducing extensive time series preprocessing and training time by an order of magnitude. Feature attribution using integrated gradients reveals that AEF provides stable baseline representations, while Sentinel-2 inputs add phenology-related detail. The results show, that integrating generalized embeddings with specialized spectral-temporal features improves predictive performance for forest mapping.

In Japan, many urban park trees are becoming large and aged, increasing the risk of structural failures caused by extreme weather events and biological deterioration. Effective management therefore requires reliable risk assessment, for which accurate tree species identification is one of the fundamental prerequisites. However, species identification still depends heavily on visual assessment by skilled professionals, posing challenges in efficiency and objectivity. This problem is particularly significant for broad-leaved trees, which exhibit high species diversity and morphological variability. In addition, labor shortages have intensified the demand for automated and reliable classification techniques. This study proposes a high-accuracy classification method for broad-leaved tree species using ground-level images captured with a commercially available RGB camera and deep learning. The proposed method extracts small local patches that capture species-specific visual features, such as leaf shape and bark texture, commonly used by professional arborists for species identification. These local features are evaluated individually using deep learning models, allowing fine-scale visual characteristics to be effectively utilized for classification. To address variability in outdoor imaging conditions, including illumination changes, shadows cast by branches and leaves, and moss attachment, multiple patches are classified independently and the results are integrated through majority voting, improving classification robustness. Experiments were conducted on seven tree species commonly found in Japanese urban parks: cherry, ginkgo, zelkova, konara oak, sawtooth oak, plane tree, and flowering dogwood. The results demonstrate that the proposed method achieves a maximum classification accuracy of approximately 95% under real-world conditions, demonstrating its effectiveness for practical urban tree management.

Geometric correction of satellite images is an essential pre-processing step for accurate geospatial analysis, but non-experts often face practical limitations because detailed sensor models and Ground Control Point data are not readily accessible. Traditional methods rely on physical sensor models or the Rational Function Model (RFM) using vendor-provided Rational Polynomial Coefficients (RPC). However, this information is often unavailable or lacks sufficient accuracy. This paper proposes a two-stage framework that utilizes AI matching technologies and open access data to automatically correct satellite images lacking georeferencing information. In Stage 1, a coarse Affine correction is executed using SuperPoint and LightGlue with an open basemap (Sentinel-2). In Stage 2, precise corresponding points are extracted through patch-based hierarchical LoFTR matching, and 3D GCPs are generated utilizing the SRTM. Subsequently, sensor-independent RPC are robustly estimated through the rpcfit library, and the final geometrically corrected image is generated through resampling.

This framework was verified by applying it to 4.8m resolution BlueBON satellite images that lack georeferencing information. In seven experimental regions with diverse geographical characteristics, an average Root Mean Square Error (RMSE) of 8.050m (1.68 pixels based on BlueBON resolution) referenced to the Sentinel-2 basemap, and an average of 9.02m (1.88 pixels) referenced to Google Maps, was achieved. This result demonstrates that it is possible to precisely correct 4.8m medium-resolution images using a 10m open basemap, providing a practical, accessible, and automated geometric correction solution for general users.

Star centroid localization accuracy fundamentally limits spacecraft attitude determination precision. Existing methods face a critical accuracy-efficiency trade-off: traditional intensity-weighted approaches achieve computational efficiency (<1 ms/star) but suffer from poor noise robustness, while Gaussian fitting and deep learning methods provide high accuracy at prohibitive computational costs. We address this fundamental limitation by developing a principled Bayesian Multi-Scale Adaptive Iteratively Weighted (BMAI) centroid localization algorithm that achieves high accuracy approaching theoretical limits while maintaining real-time computational efficiency. The algorithm integrates four key technical contributions: (1) SNR-adaptive window extraction with robust threshold estimation, (2) regularized iteratively weighted framework with proven convergence properties, (3) multi-scale fusion with SNR-dependent weighting, and (4) gradient-based refinement to mitigate systematic bias. Rigorous theoretical analysis establishes convergence guarantees, derives error bounds, and evaluates Cramér-Rao Lower Bound (CRLB) efficiency. Comprehensive evaluation on 16,500 synthetic star images across six diverse imaging scenarios demonstrates that under high-SNR conditions (SNR >25, n=2,000), BMAI achieves mean RMSE of 0.0120 pixels (95% CI: [0.0116, 0.0124] pixels), representing a 98.6% relative improvement over intensity-weighted centroiding (0.857 pixels), 35.8% improvement over Gaussian fitting (0.0187 pixels) and 95.3% improvement over CNN methods(0.2566 pixels). The algorithm maintains computational efficiency of 0.89ms per star—8.7× faster than Gaussian fitting—while achieving CRLB efficiency of 79.2%. Robustness analysis demonstrates stable performance across SNR range 3-100 with graceful degradation under challenging conditions. The BMAI algorithm fundamentally resolves the accuracy-efficiency trade-off in star centroid localization through principled Bayesian inference and multi-scale processing.

This paper presents a systematic investigation of the PhaseOne native IIQ format for drone and aerial cameras (in particular the recent iXM-RS250 and the iMX-GS120), focusing on the influence of different compression levels on geometric, radiometric and computational aspects of the photogrammetry pipeline. The aim of the presented research and experiments is to demonstrate the actual quality of these (compressed) images for photogrammetric purposes.

Small format multi-head cameras are becoming available and can be flown on light drones to provide simple access to oblique and nadir views of built-up areas. A number of missions with different parameters (flying height, etc.) are investigated to understand the trade-offs in applying those sensors and question the established accuracy laws. We investigate and quantify the ability to completely cover the facades using those sensors in the different scenarios.

In recent years, commercial small optical satellites, e.g., Skysat, BlackSky, and PlanetScope, have become widely used for a variety of Earth remote sensing applications, providing high-resolution images with sub-meter resolution. They are operated in a constellation of multiple satellites, which compensates for the spatial and temporal limitations of traditional satellite observations. Moreover, their data acquired during stereo viewing have been experimentally used to generate digital surface models (DSMs). The CE-SAT-IE is a small optical satellite developed by Japan’s commercial company Canon Electronics Inc. (CE) and was launched on 17 February 2024, by Japan Aerospace Exploration Agency’s (JAXA’s) H3 launch vehicle test flight no.2. It is equipped with an optical frame sensor derived from a commercial off-the-shelf (COTS) camera Canon EOS R5. The ground sampling distance (GSD) is 0.8 m with a scene size of 6.5 km × 4.3 km. The calibration and validation of the sensor are being conducted in collaboration between CE and JAXA, drawing on JAXA’s extensive experience with past satellites. The geometric and radiometric performance of the sensor is analysed in detail, and the results will be used for its subsequent mission, which may involve a constellation for stereo observation to generate high-quality DSMs. This paper reports initial results for geometric calibration and validation of the sensor using ground control points (GCPs) and the experimental generation of DSMs from stereo observation images using the calibrated parameters.

This paper presents a novel hybrid calibration pipeline that jointly estimates the spatial and temporal alignment between a handheld laser scanner and a smartphone camera without any hardware synchronization. The method combines deep-learning-based target detection with classical geometric calibration using 2D-3D correspondences derived from black and white hourglass planar targets. Target centers are precisely localized in both the RGB images and the 3D point cloud using a symmetric templatematching scheme, enabling robust solution of the perspective-n-point (PnP) problem for spatial calibration.

To address the lack of hardware synchronization, we introduce a temporal calibration method that exploits geometric correspondences between rendered intensity images and camera frames.

On a Lixel L2 Pro scanner with a Huawei P20 Pro camera, the pipeline achieves a median Reprojection error of 0.76 px for static calibration and 2.19 px across 91 dynamic evaluations. The approach enables accurate image-pointcloud fusion for scanners without syncronisation interfaces and provides a foundation for colorization, image analysis, and redensification of laser data.

Accurate large-scale dense 3D reconstruction is fundamental for geospatial mapping, robotics, and autonomous navigation. Conventional photogrammetric workflows can reconstruct environments from ground-level imagery but often suffer from cumulative drift over kilometer-scale trajectories and require extensive calibration. Recent advances in feed-forward 3D reconstruction, notably the Visual Geometry Grounded Transformer (VGGT), have demonstrated the ability to generate dense point clouds directly from RGB images without explicit optimization. VGGT jointly estimates camera poses, depth, and dense geometry from multiple uncalibrated frames in a single forward pass. However, its scalability is limited by two factors: (1) the lack of absolute metric scale and (2) high GPU memory demands.

Many national mapping agencies (e.g., USGS, IGN, Ordnance Survey) have released Airborne Laser Scanning (ALS) datasets covering vast urban and rural areas. These high-quality aerial point clouds provide globally consistent, metrically referenced data that can serve as external constraints for ground-level reconstructions. Building upon this opportunity, we propose VGGT-ALS, a framework that leverages open ALS point clouds to enable VGGT-based systems to produce large-scale, metrically accurate dense maps from mobile mapping imagery.

High-precision vehicle localization in GNSS-denied urban areas requires alternatives to costly HD maps. In this paper, we present a novel framework for feature extraction and benchmark generation to enable high-precision localization using abstract LoD2/DTM maps as a replacement for HD maps. Our first contribution, a semantic-geometric pipeline, processes dense LiDAR and camera data to extract map primitives. This is accomplished by a RANSAC-fitted ground plane extraction step, followed by a semantic filter that discards dynamic objects. Finally, geometric clustering (HDBSCAN) and RANSAC plane fitting isolate large-scale vertical facades. Our second contribution, a multi-stage GT generation framework, resolves annotation ambiguity using a Human-In-The-Loop (HITL) system. A robust 2D pose is computed by finding the geometric median of bootstrapped transformation samples on the SE(2) manifold, which is then refined to a 6-Degree-of-Freedom pose via point-to-plane ICP, before being validated by a human for a final check. We evaluated our feature extraction pipeline against the generated benchmark, achieving 95.04% precision and 83.74% recall. An analysis of this performance shows the pipeline correctly rejects small, ambiguous features while achieving high recall on all large, stable features, proving its suitability for a robust localization filter.

As civil infrastructure in South Korea ages, the demand for systematic monitoring has grown. While Unmanned Aerial Vehicles (UAVs) provide a safer alternative to manual inspections, Global Navigation Satellite System (GNSS) signal degradation beneath bridge structures remains a critical barrier to autonomous flight. Unlike existing hardware-centric or SLAM-based solutions that require high costs and computational overhead, this study proposes a robust, algorithm-based path planning methodology using a 3D spatial grid framework.

Two strategies were evaluated through field tests at the Bukhangang Bridge: the Photography Point Method (PPA) and the GNSS Non-Shadowing Area Method (WPS-GNSA). Results demonstrated that WPS-GNSA significantly enhances signal reliability by strategically positioning waypoints outside electromagnetic shadow zones. At an optimal 19-meter separation distance, WPS-GNSA maintained GNSS Level 5 connectivity—the threshold for recommended autonomous flight—for 9.4% of the duration and Level 4 or higher for 79.7%, whereas the PPA peaked at Level 4.

These findings indicate that WPS-GNSA enables reliable autonomous inspections using standard commercial drones without specialized hardware modifications. While the current model relies on pre-existing digital blueprints, future research will integrate AI-based navigation and real-time environment perception to enhance scalability and adapt to dynamic obstacles in complex infrastructure environments.

Accurate and reliable positioning is fundamental for the development of a wide number of applications. Despite in most of the regular working conditions the use of a GNSS receiver is sufficient for properly solving the problem, determining a reliable solution in challenging conditions can be difficult. In such conditions, exploiting the information shared by different sensors and platforms can be useful for reliably determining the platforms' positions. In this work, both ground and aerial platforms are considered: each platform is assumed to be provided with communication capabilities, which can be exploited to share its knowledge. Since GNSS positioning is usually less effective at ground level than on a flying platform, the aerial platforms are assumed to be provided with good GNSS-based positioning information. Instead, GNSS is assumed to be unavailable to the ground vehicles, which, instead, can use LiDAR/visual odometry for dead-reckoning positioning, UWB inter-platform ranging for relative positioning, and camera-based positions, provided by aerial platforms, for assessing their georeferenced positions. This work focuses on assessing the positioning performance when exploiting vision-based information about the georeferenced ground vehicle positions from a camera mounted on a UAV. The camera acquired oblique views of the scene while moving over the case study area during the test. YOLO was used to detect cars from the image frames and the vehicle coordinates have been extracted from 3D reconstructions obtained from the MoGe-2 network. Average errors at meter level on the determined georeferenced coordinates were obtained when combining UWB vehicle-to-vehicle ranges with MoGe-2 reconstructions.

Reliable positioning and mapping in dense urban environments remain challenging due to signal blockage, multipath, and dynamic scenes. Progress on multi-sensor integrated positioning and visual/lidar SLAM has been driven by open datasets, yet most existing resources are either perception-centric with limited raw navigation data, focused on controlled environments, or built around outdated software platforms and/or data formats. In this paper, we present the Melbourne Multi-Sensor Urban Positioning and Mapping Dataset, a new resource targeting urban vehicle navigation and mapping tasks. The dataset was collected using a custom mobile mapping platform equipped with a tactical-grade INS, a survey-grade Leica GNSS receiver, a low-cost UBLOX GNSS receiver, a high-resolution Ouster OS1 128 lidar, and four industrial FLIR cameras providing 360° coverage. Seven data collection trips were recorded on dynamic streets in several inner suburbs of Melbourne, including multiple closed loops and a repeated route with day–night variation. For better compatibility and future-proofing, all raw data are provided as standard ROS2 message streams in MCAP format, complemented by commonly used individual formats and GNSS products for multi-sensor integrations. We benchmark three GNSS--based positioning packages (RTKLib, Net_Diff and Ginan) and four state-of-the-art lidar(-inertial) odometry/SLAM methods (FAST-LIO2, KISS-ICP, KISS-SLAM and PIN-SLAM), demonstrating the applicability and compatibility of our dataset for modern positioning and mapping software pipelines. The dataset is designed as a robust, ROS2-native testbed for research on GNSS/IMU/lidar/camera fusion for the testing and validation of vehicle positioning and mapping in urban environments, which is available open-source at https://github.com/zjjdes/melbourne_dataset.

Accurate and robust localization is essential for autonomous vehicles and mobile robots operating in complex, dynamic environments. However, existing learning-based LiDAR-inertial odometry (LIO) methods typically rely on simple weighted fusion or purely global attention, which may not fully exploit cross-modal complementarity. In this paper, we propose CMLGF-LIO, a cross-modal local-global fusion framework that improves LIO accuracy and robustness. At the local level, we design a Local Split-Attention (LSA) module that injects IMU-derived motion priors into local LiDAR feature groups and adaptively allocates attention weights, suppressing redundant information while preserving discriminative local geometry for fine-grained fusion. At the global level, we introduce a Global MLP-Mixer (GMM) module that aligns LiDAR and IMU token sequences and models global cross-modal interactions using an MLP-Mixer backbone. Experiments demonstrate that CMLGF-LIO is more robust than learning-based baselines under challenging conditions, and ablation studies validate the effectiveness of the proposed local-global fusion strategy.

This paper presents a low-cost mobile mapping system built on a consumer vehicle (2026 Tesla Model Y) equipped with a roof-mounted Velodyne VLP-16 LiDAR and three post-processed kinematic (PPK) GNSS receivers. A processing pipeline was developed that fuses scan-to-scan LiDAR registration via KISS-ICP with multi-receiver PPK trajectories to produce georeferenced 3D point clouds. The system was tested on a 2 km loop on the Ohio State University campus. KISS-ICP achieved a registration fitness of 1.0 across all 3,400+ frames, producing locally crisp point clouds. A yaw-only Procrustes alignment followed by interactive 7-parameter refinement maps the SLAM trajectory into an East-North-Up geodetic frame. We document the complete pipeline architecture, including automated GPS time synchronization, multi-receiver vehicle pose estimation, and a streaming LAS export capable of handling 70+ million points. We systematically evaluate ten post-hoc trajectory correction strategies and identify a fundamental trade-off between inter-frame consistency (point cloud crispness) and absolute geodetic accuracy. The primary unresolved challenge is a staircase artifact caused by ~40 m of accumulated SLAM drift over the loop, which cannot be corrected without degrading local registration quality. We conclude that loop closure detection and pose graph optimization within the SLAM pipeline are necessary to resolve this tension and outline a path toward survey-grade mobile mapping from consumer vehicle platforms.

Estimating carbon stock is important for understanding ecosystem dynamics and mitigating climate change. However, biomass mapping in boreal forests faces challenges due to harsh conditions and limited ground truth data for large scale studies. This study presents a parametric model for accurate biomass estimation in the Acadia and Taiga Forest using GEDI Level 4A, Sentinel-2, and SRTM DEM data. We integrated these datasets, and developed the parametric model consisting of spectral bands, vegetation indices, and topographic information with regression techniques, Random Forest and K-nearest neighbour. Results showcase performance of the parametric model with relative weights of variables for accurate Aboveground Biomass Density (AGBD) predictions for the two forest sites. With an average RMSE ranging between 9 Mg/ha to 31 Mg/ha and R^2 values of 0.54 to 0.60, the study reveals the importance of variables like slope, aspect and specific vegetation indices along with raw bands of Sentinel-2 data. Results also demonstrate potential and accuracy limitations of the proposed model with for biomass estimation with high resolution open-source satellite data without ground control. Further research include assessing the model robustness across diverse ecosystems and geographical settings, contributing to sustainable resource management practices.

Accurate estimation of aboveground biomass (AGB) is essential for quantifying forest carbon stocks, monitoring ecosystem change, and supporting greenhouse-gas reporting frameworks. While field measurements remain the benchmark, their limited spatial coverage has driven increasing reliance on remote sensing. Existing global AGB products such as ESA CCI Biomass and GEDI represent major advances but still suffer from signal saturation, sparse sampling, and limited ability to resolve fine-scale structural variation, highlighting an ongoing gap in effectively fusing information from different sensors. Recent studies combining Sentinel-1 (S1) SAR and Sentinel-2 (S2) multispectral imagery have demonstrated improvement in biomass modelling, with machine learning and deep learning achieving R² values from 0.57 to 0.80. However, most work focuses on tropical or broadleaf forests, leaving boreal mixedwood systems underrepresented, and models often rely on short-term composites that overlook multi-year temporal dynamics important for distinguishing long-term growth from seasonal phenology. This study addresses these gaps by developing a Transformer-based deep learning framework that integrates multi-temporal S1 and S2 time series and incorporates Growth & Yield (G&Y) variables as temporal constraints. By leveraging complementary radar–optical interactions and long-range temporal dependencies, the model is designed to reduce signal saturation, enhance structural sensitivity, and improve generalizability across heterogeneous boreal forest conditions.

Fuel Moisture Content (FMC) serves as a vital indicator for monitoring vegetation health and predicting wildfire risk. While existing approaches have largely emphasized temporal variations in FMC, they frequently overlook the inherent spatial clustering patterns of vegetation, leading to compromised spatial prediction accuracy. To overcome this limitation, we introduce a Transformer-based Spatio-Temporal Estimation Framework (TSTEF) that preserves sensitivity to temporal dynamics while incorporating spatial aggregation mechanisms to achieve robust and spatiotemporally consistent FMC estimates. The framework combines spatial autocorrelation analysis with gated recurrent unit (GRU)-based temporal modeling to effectively capture spatiotemporal dependencies, and utilizes Triangular Topology Aggregation Optimization (TTAO) for hyperparameter calibration. The proposed framework was validated using Sentinel-1/2 imagery and MODIS products in California, USA, where it demonstrated: (1) outstanding performance with an average R² > 0.8, MAE < 9%, and relative RMSE of 12.35%; (2) strong agreement between estimated FMC distributions and ground observations, with wildfire burned areas significantly expanding when FMC fell below the 120% critical threshold; and (3) excellent generalization ability during cross-regional validation, achieving relative RMSE values of 20.46% in France, 25.62% in Spain, and 20.76% in Colorado. This study provides a reliable analytical framework for wildfire risk early warning and contributes meaningful insights for ecosystem management amid global environmental change.

In the United States (U.S.), aggressive suppression policies in the 20th century reduced wildfires in the short term, but accumulated fuels contributed to increased wildfire risk in the long term. Land managers are slowly reintroducing the use of fuel treatments, including prescribed (Rx) fires, to remove fuels and mitigate future wildfires. To date, efforts to systematically quantify the efficacy of fuel treatments for wildfire mitigation have been limited in spatial and temporal scope or rely on proxies, such as low-intensity wildfires. Here we use the 30-m Harmonized Landsat and Sentinel-2 (HLS) dataset to analyze timeseries of vegetation “greenness” indices, such as the Normalized Burn Ratio (NBR), with observations up to every 2-3 days. We apply a causal inference approach, difference-in-differences (DID), on HLS-derived timeseries of NBR to compare outcomes in treated and surrounding control areas in three different time phases: 1) post-treatment fuel reduction, 2) wildfire-induced burn severity, and 3) post-wildfire vegetation recovery. As a case study, we targeted 37 Rx fires that preceded and intersected the 2024 Park Fire, a large wildfire in northern California. We show statistical evidence that Rx fires reduce fuel loads (12 Rx fires), wildfire burn severity (12 Rx fires), and post-wildfire vegetation recovery (14 Rx fires). Our approach requires only the spatial footprint and timing of the fuel treatments, thus enabling regional to nationwide analyses using a large number of fuel treatments to quantify the general efficacy of fuel treatments across a variety of treatment types, fuel types, and topography.

Ecosystem extent has been selected as an indicator of biodiversity under the Kunming-Montreal Global Biodiversity Framework. With wildfires on the rise in Canada and around the World there is a need to understand their impact on ecosystem extent for Framework reporting requirements. In Canada’s Arctic and Sub-arctic the growing season is much shorter than at lower latitudes, resulting in few cloud-free optical images and making it a challenge to monitor the impacts of fire and recovery at fine temporal resolutions. Synthetic Aperture Radar (SAR), particularly winter coherence, can be an indicator of ecosystem extent in the sub-arctic because burned areas will exhibit patterns that reflect more dynamic freeze-thaw cycles in the winter period than non-burned areas. Winter coherence and phase was calculated for 39 Sentinel 1 images over the time periods of 2017-2018, 2018-2019, 2019-2020, 2020-2021 in Northern Manitoba, Canada. When relating these values to known fire areas it was found that there is a distinct difference between areas recently burned as opposed to those burned further in the past. These data will be used as predictor variables in a Random Forest ecosystem classifier with outputs of overall accuracy and Shapley feature importance assessed.

Accurate and frequently updated land cover maps are vital for various scientific communities, as well as for public and regional authorities, supporting decision-making, planning, sustainable development, and natural resources management. Regular monitoring and mapping also play a crucial role for agricultural areas, particularly considering the projected population growth and shifting dietary patterns in many of the fastest growing regions of the world, that pose significant challenges for humanity. Over the past decade, the availability of Sentinel-1 and Sentinel-2 data has significantly increased the potential for high spatial resolution land cover mapping using dense time series. However, mapping croplands and distinguishing between crop types remains a more complex task, often requiring data of higher spatial, spectral, and temporal resolution. In this context, this study aims to evaluate the synergistic use of multi-temporal data from Sentinel-1, Sentinel-2, and EnMAP data for detailed land cover and crop type mapping in agriculural regions of western Greece.

Urban development in the Greater Toronto Area faces significant challenges because zoning abbreviations and terminology vary widely between municipalities. This provides the background and motivation for the study, as labels such as “R2” in Toronto and “R2 S” in Markham appear similar yet represent different permissions and development standards, creating confusion and slowing planning workflows in a region with growing housing pressures. The problem addressed in this research is the absence of a unified, machine-readable framework that standardizes zoning terminology across municipalities, which limits automated compliance checking, GIS integration, and cross municipal comparison. The objective of this work is to create Zonology, an ontology-based framework that harmonizes zoning abbreviations, permitted land uses, and development standards, beginning with the City of Toronto and the City of Markham. The methodology follows the Linked Open Terms approach, using the Web Ontology Language to encode zoning by laws, land use categories, development standards, and spatial relationships. The model is evaluated through reasoning tasks, competency questions, and semantic alignment tests to ensure clarity, consistency, and interoperability. The results show that Zonology successfully aligns more than sixty zoning categories and over one hundred fifty land use permissions, enabling consistent semantic interpretation and cross municipal queries. The overall significance of this work is that the ontology improves regulatory clarity, strengthens data driven planning, and provides a scalable foundation for harmonized zoning governance across the Greater Toronto Area.

Urban systems are increasingly complex, interconnected, and dynamic, yet most geospatial data models continue to represent them as static geometric layers with limited support for explicit relationships and semantics. This restricts advanced spatial reasoning, network analysis, and AI-driven applications.

This paper introduces GeoGraphJSON, a lightweight semantic data model that extends GeoJSON by integrating spatial geometry with graph-based connectivity. The framework represents spatial entities as nodes and explicitly encodes relationships as typed edges, enabling unified representation of geometry, topology, and semantics within a single interoperable structure. A hierarchical Unique Identifier (UID) system ensures consistent lineage and cross-layer integration across administrative, transportation, and urban asset datasets.

The approach is validated using a large-scale urban dataset from Riyadh, comprising over 10,000 nodes and 13,000 edges. Graph-based analysis demonstrates realistic spatial patterns, including right-skewed degree distribution, strong network connectivity, and identifiable community structures. These results highlight the ability of GeoGraphJSON to capture hierarchical organization and functional relationships while supporting efficient analytical workflows.

By bridging geometry-centric GIS models and graph-based approaches, GeoGraphJSON provides a scalable foundation for urban analytics, digital twins, and GeoAI applications, enabling geospatial systems to evolve from static representations toward intelligent, relationship-aware spatial models.

Urban morphology quantitatively reveals how distinct historical and functional drivers shape city form. This study employs a computational morphometric approach using the Momepy library to analyze and compare the urban structures of Cairo, Egypt, and Makkah, Saudi Arabia. These cities represent paradigmatic cases: Cairo exemplifies long-term, organic layering, while Makkah demonstrates rapid, purpose-driven transformation for religious pilgrimage. We calculated key metrics—including tessellation area, convexity, elongation, equivalent rectangular index, and edge betweenness centrality—for building footprints and street networks sourced from OpenStreetMap. Results show Cairo possesses a heterogeneous, polycentric fabric with complex plot shapes and a distributed street network, reflecting its layered history. Conversely, Makkah exhibits a more monocentric, consolidated form with standardized building geometries and a hierarchical street network channeling movement toward its core. The findings demonstrate that quantitative morphology effectively captures how Cairo's organic evolution and Makkah's centralized planning produce fundamentally different, yet equally revealing, urban spatial structures, offering a replicable framework for cross-city analysis in the region

Urban illumination, as captured through Nighttime Light (NTL) data, serves as a powerful indicator of human activity, infrastructure development, and socioeconomic progress in rapidly growing cities. However, previous studies on Delhi have largely focused on temporal NTL trends without integrating multi-year statistical and spatial analyses to reveal underlying urban and socioeconomic dynamics. This study investigates the spatiotemporal dynamics of urban illumination and development over Delhi using VIIRS Day/Night Band (DNB) NTL data for the years 2015, 2020, and 2025. NTL intensity was used as a proxy for urbanization and socioeconomic activity. Monthly composite datasets for January of each year were processed, clipped to the Delhi administrative boundary, and analyzed using statistical, temporal, and correlation-based methods. The results revealed a slight decline in mean NTL intensity from 26.34 in 2015 to 24.95 in 2025, indicating stabilization in overall light emissions may be due to the adoption of energy-efficient technologies. However, the maximum and range values increased markedly (166.85 to 228.04), signifying intensified illumination in high-activity commercial and infrastructural zones. Temporal change analysis showed balanced positive and negative illumination shifts, with over 50% of pixels exhibiting moderate growth during 2020–2025. Strong Pearson and Spearman correlations (r = 0.83–0.92; ρ = 0.910.95) confirmed the temporal consistency of illumination distribution. The socioeconomic assessment highlighted spatial disparities in light intensity might be corresponding to varying economic activity levels. Overall, the study demonstrates that VIIRS-derived NTL data provide an effective and robust approach for monitoring urban growth, socioeconomic variability, and sustainable lighting transitions in metropolitan environments.

The research investigates artificial intelligence as a device for the automatic interpretation of landscape, reframing representation not as a neutral reproduction but as a cognitive operation in which perception, description, and evaluation converge. Moving from the assumption that landscape is not an objective given but a culturally and perceptually constructed form, the study proposes a fully data-driven methodology based on geolocated images. Through a systematic grid sampling, street-level panoramic views are collected and processed within a multimodal pipeline integrating visual analysis, language models, and multi-agent evaluation. Images are first translated into textual descriptions and semantically clustered, allowing territorial classes to emerge from the data rather than from predefined taxonomies. In parallel, a simulated cognitive framework, structured through four agent profiles, produces evaluative scores and textual judgments, later analysed through sentiment detection. The integration of these layers generates a georeferenced dataset from which a perceptual cartography of the territory is constructed. Applied to the urban context of San Giustino (Italy), the method reveals a continuous gradient from dense urban cores to rural landscapes, while exposing differentiated perceptual readings across observer profiles. Within this framework, artificial intelligence does not replace human interpretation; it operates as an epistemic extension, transforming the landscape into a distributed field of comparable perceptions, where representation becomes a computable form of shared knowledge.

This study focuses on bridging usability information between data providers and data users through standardized metadata. By further integrating standardized metadata with geographic information system operation design, the operations gain automated and awareness capabilities, allowing usability information based on data specifications and quality considerations to be incorporated into relevant processes, thereby avoiding erroneous decisions. The research references international standards such as ISO 19115 and ISO 19157 to meet the requirements of open geographic information technologies.

This invited talk presents Digital Building Analysis (DBA), a unified framework for intelligent, cloud-based building-scale reconstruction and analysis. Building on our prior advances in Gaussian Splatting for photorealistic 3D scene generation and the Gaussian Building Mesh (GBM) framework for accurate mesh extraction, DBA introduces a new layer of integration between cloud mapping and artificial intelligence. The system connects directly to Google Maps Platform APIs to retrieve geospatial data, imagery, and elevation models from a building’s address or coordinates, while employing Gaussian Splatting to reconstruct high-fidelity 3D models from multi-view imagery. This combination enables seamless digital twin creation without ground-based measurements or proprietary datasets and can be operated through natural language queries, allowing users to simply describe a location or request a building analysis conversationally.

The key component of DBA leverages multi-agent large language models (LLMs) for both natural language interfacing and data interpretation. These models autonomously generate Google Maps API calls, interpret retrieved imagery, extract visual features, and compose semantic building descriptions. Working in tandem, the agents merge 3D geometry, visual realism, and semantic understanding into a single automated process.

Together, these innovations mark a major step forward in Canada’s AI-enhanced remote sensing research, enabling interactive, query-driven urban analytics and advancing the next generation of intelligent digital twins for sustainable urban development.

3D building models are powerful visual tools, typically generated with well-established image-matching or LiDAR methods. However, they do not capture the view-dependent colour characteristics possible with Gaussian splatting. Despite the visual potential of Gaussian splatting, there is limited knowledge on its spatial accuracy and influencing factors, particularly for buildings. To address this gap, a two-building dataset was collected with terrestrial laser scans, images, phone LiDAR, and target points, and the visual and spatial effects of numerous factors were analyzed. These factors included the source and quality of the input camera poses and point cloud, the number of images and training iterations, and the Gaussian splat method. Gaussian splats were trained from open source and commercial image-based reconstruction methods, COLMAP and Pix4D, and phone LiDAR reconstructions. Applying Gaussian splatting to these inputs had minimal impact on the target points and the overall structure of the buildings, but the positions of Gaussians deviated from the initial point cloud, particularly before 15,000 iterations, resulting in more floaters and lower spatial accuracy. Image-based reconstruction methods outperformed phone LiDAR methods on visual and spatial metrics. Cleaning COLMAP point clouds considerably decreased Gaussian floaters, while downsampling input point clouds increased the percentage of floaters and yielded similar visual results. 2D Gaussian splatting provided geometric constraints, removing some floaters, but sacrificed visual quality. Increasing the number of images to three loops around the building improved visual and spatial results. Overall, the spatial accuracy of building Gaussian splatting was heavily dependent on the factors studied.

Urban Digital Twins (UDT) are vital tools for smart city development, enabling data-driven management and analysis of urban infrastructure (Sabri and Witte, 2023). A persistent challenge in realizing the potential of UDTs is the interoperability of disparate geospatial datasets, particularly camera imagery and sensor data, requiring precise synchronization, georeferencing, and integration. Existing implementations often

rely on costly, proprietary hardware, limiting scalability and adoption, especially for organizations constrained by limited budgets (Thakkar et al., 2025). This research addresses the need to develop a cost-effective, standardized framework to capture, integrate, and standardize camera imagery and geospatial metadata for Machine Learning (ML)-driven analysis within spatially enabled UDTs.

Roof material identification is a critical component of energy-aware 3D city modeling, supporting applications such as thermal analysis, climate resilience, and digital twins. Traditional approaches relying solely on aerial imagery struggle with shadows, low contrast, and spectrally similar roof materials. This study introduces a dual-branch deep learning framework that combines high-resolution aerial orthoimages with GoPro-based street-view imagery to overcome these limitation and improve roof material classification.

The aerial branch employs a ResNet-18 model fine-tuned on 120 manually labelled roof samples in New Brunswick, Canada, covering four material classes: asphalt, metal, membrane, and gravel. The street-view branch utilizes GoPro field-survey images, where roof regions are extracted using the Segment Anything Model (SAM) before classification with a second ResNet-18. Although street-view imagery captures only materials visible from ground level, it offers rich textural information that complement nadir imagery.

Because the two modalities are unpaired, fusion is performed at the decision level using learnable weights to combine the softmax probabilities of both branches. Experimental results show that street-view imagery achieves 90.9% accuracy, outperforming aerial imagery alone (77.8%). The combined bimodal framework leverages complementary modality strengths, resulting in improved detection performance for all roof material classes.

This study investigates the comparative performance of state-of-the-art 2D and 3D feature-detection models applied to multimodal airborne datasets for digital-twin generation. Using RGB, LiDAR, and nighttime thermal imagery collected over the University of New Brunswick’s Fredericton campus, a fused RGB–LiDAR–thermal point cloud was created to support building-scale analysis of energy-relevant features, specifically windows and doors. Three 2D object-detection models Faster R-CNN, Mask R-CNN, and YOLOv8 were applied to both RGB and thermally registered imagery, incorporating phase-congruency-based alignment to address differences in sensor geometry and spatial resolution.

Complementing the 2D analysis, three 3D semantic-segmentation models KPConv, PointCNN, and RandLA-Net were implemented to evaluate geometry-driven, order-aware, and scalable point-cloud classification strategies using multimodal attributes. The dataset was divided into 70% training and 30% testing, and evaluated using standard metrics such as accuracy, mean Intersection-over-Union, and per-class F1 score. Preliminary results for the 2D methods have been realased in the abstract, with further evaluation of all models currently underway.

The objective of this work is to establish a unified framework for understanding how 2D and 3D feature-detection approaches perform under low-light and thermally dominant conditions, where conventional RGB-based workflows often fail. The outcomes of this study will support improved digital-twin development for building-energy diagnostics and contribute to future thermal-efficiency modeling workflows in partnership with the National Research Council of Canada.

The Terrasolid software suite provides an automated and scalable framework for large-scale LiDAR data processing, widely adopted in both national and private forest inventories. Its unified processing pipeline covers all essential steps—from point cloud import and georeferencing to ground classification, object detection, tree segmentation, and computation of individual-tree attributes such as diameter at breast height (DBH), height, volume, and tree species. Georeferencing is initially performed in TerraScan using signal markers or automatically detected tree trunks, with optional refinement in TerraMatch, which corrects angular misalignments between flight lines.

Following object classification, individual trees are extracted from points labeled as trees. The semi-manual Group Inspection tools support efficient correction of segmentation errors, such as merged or over-segmented trees, after which stem-wise metrics are automatically updated. These conventional modules rely on optimized algorithms capable of processing hundreds of millions of points within minutes.

A recent innovation, the Tree Species tool, introduces one of the first AI-based extensions within Terrasolid software. It employs a machine learning approach that integrates 2D raster-based features with 3D point cloud descriptors to achieve accurate tree species identification. Validation was conducted using the FOR-species20K dataset, comprising 33 species collected worldwide. Among several tested classifiers, the Histogram Gradient Boosting Classifier (HGBC) achieved the highest accuracy. To mitigate class imbalance, multiple side-view rasterizations and SVM-SMOTE oversampling were applied, significantly improving the separability of underrepresented species and overall classification robustness.

Traditional aboveground biomass estimation for forested areas relies on allometric equations (Návar, 2009), which use input variables such as diameter at breast height (DBH), tree height, and tree species or broader taxonomic group. Although allometric equations can estimate the biomass of individual trees, and stand-level equations exist for larger scales, they often require extensive field data, making them less suitable for densely clustered or remote forests. However, satellite images provide increasingly detailed global observations of forested areas, and spaceborne lidar data like GEDI (Duncanson et al., 2022) provide accurate measurements for canopy height across different ecozones worldwide. In recent years, foundation models (FMs) inspired by large language models (Vaswani et al., 2017) have become the new paradigm to leverage large amounts of unlabelled data through self-supervised pre-training and have shown capacity to benefit multiple downstream tasks. In this work, we adopt the Granite foundation model (Muszynski et al., 2024) as a baseline to improve aboveground biomass estimation on different satellite data, using the Mixedwood Plains Ecozone (MPE) as a case study. We also explore adding temporal, geospatial, and spatiotemporal features and validate the proposed spatiotemporal foundation model with field sampling plots.

Accurate detection and mapping of individual trees and their species are vital for sustainable forest management. Traditional field-based inventories remain the golden standard in forest monitoring, but are increasingly overwhelmed by temporal, spatial and accessibility constraints. Remote sensing offers faster, repeatable, and high-resolution data that complement and scale beyond field inventories. However, species-level detection remains difficult due to overlapping crowns, and spatial mismatches between crowns and trunks. Deep learning (DL) methods, particularly convolutional neural networks, have advanced crown delineation by automatically learning spatial and spectral patterns from imagery. Yet, their success depends heavily on dataset quality, class balance, and diversity. To address this, we applied a DL object detection framework for tree crown and species detection in Swiss forests and evaluate how dataset composition and training strategies influence accuracy and generalization. We test three dataset configurations: (1) an unbalanced masked dataset, (2) a class-balanced masked dataset, and (3) a mixed dataset combining masked and unmasked imagery. Results show that class balancing improved accuracy for both dominant and minority species, while mixed data enhances generalization.

High-resolution airborne sensing enables joint mapping of urban infrastructure and forest composition at ecological scales. This study presents a self-supervised segmentation framework that fuses 0.5 m Near-Infrared + RGB (NIRGB) orthophotography from the Ontario Imagery Program (2013–2026) with Canopy-height models (CHM) derived from the Ontario Elevation Mapping Program (8–10 pulses m⁻², 5–10 cm vertical accuracy). Imagery was collected during the leaf-off season, providing strong spectral–structural contrast between evergreen and deciduous crowns, to produce high-fidelity land- cover segmentations that differentiate vegetation functional types and built structures as a prerequisite for tree-level biomass and carbon-stock estimation.

Spatially accurate forestry road networks are essential for effective forestry operations, sustainable resource management, and conservation. Current forestry road databases in Ontario have significant location errors due to limitations and human errors associated with conventional road delineation approaches such as GPS-based field surveys and photointerpretation. A previously developed algorithm, which used airborne laser scanning (ALS) data, successfully corrected road locations in Quebec. However, its design limited its application in other landscapes, ALS instruments, and road construction and maintenance practices. This study advances that algorithm by integrating a deep learning component to improve its robustness and scalability for diverse forest conditions. A hybrid workflow combines the original friction-based conductivity surface with a road probability surface generated by an Attention Residual U-Net model trained on 11 LiDAR-derived features using road segments from five forest sites in Quebec. The enhanced workflow was applied to two forest management units in Ontario: Nipissing and Dryden. The results showed significant improvement in road alignment when compared to the existing provincial data and the outputs from the earlier automated approach. The deep learning-enhanced algorithm lowered mean positional error by 78% (from 9.36 m to 2.07 m) and increased the proportion of road centerline points within 3 m of the reference from 66.7% to 87.2%. These improved centerline accuracies will further support a scalable tool for rapid and accurate forestry road network mapping, which in turn will aid sustainable forest management and conservation planning at both provincial and national scales.

Monitoring tree health is essential for detecting early signs of stress, defoliation, and potential mortality, supporting effective forest management, ecosystem conservation, and early warning systems. Advances in deep learning have enabled automated analysis of trees in remote sensing imagery through object detection methods that leverage both spectral and spatial information. However, assessing tree defoliation remains challenging, as subtle differences between defoliation levels make accurate classification difficult. To address this, we propose the hybrid ResNet-Swin Transformer, an object detection architecture built on a Faster R-CNN framework, incorporating a fused ResNet and Swin Transformer backbone with attention-based feature fusion. This design captures rich, multiscale representations by combining convolutional and transformer-based features and progressively refines them through channel-wise attention blocks for robust detection and classification. The architecture was evaluated on a very high-resolution aerial dataset from Switzerland, partially annotated with five classes: Conifer (healthy), Conifer (defoliated), Broadleaf (healthy), Broadleaf (defoliated) and Dead. Comparative experiments with state-of-the-art object detection and classification methods demonstrate that the proposed approach achieves higher accuracy and robustness, highlighting its potential for precise and reliable automated tree health monitoring.

Accurate vegetation mapping in complex urban environments is essential for ecological monitoring, biodiversity assessment, and sustainable park management. However, fine-grained vegetation segmentation remains challenging because of the high diversity of plant species, overlapping canopies, and the interference of artificial objects. To address these challenges, a deeply supervised parallel architecture based on the SegFormer backbone was proposed in this paper. The model incorporated a SegFormer-ASPP-low-level (SAL) head, which fused high-level semantic representations, multi-scale contextual information, and low-level spatial details through a parallel decoding mechanism. Two auxiliary heads, a pyramid pooling module (PSP) and a fully convolutional network (FCN), were added to provide deep supervision and improve the recognition of blurred boundaries and rare categories. High-resolution UAV imagery was used to perform fine-grained semantic segmentation of 17 vegetation categories. The dataset included multiple tree species as well as non-tree classes such as Nelumbo sp. (lotus) and dead trees. Experimental results showed that our model achieved a mean intersection over union (mIoU) of 73.57%, outperforming architectures such as SegFormer-b1, DeepLab v3+, ConvNeXt and SCTNet. Visual analysis further demonstrated the model's robustness in complex urban park scenes, showing superior boundary delineation, improved recognition of small and spectrally similar species, and resilience to interference from artificial objects like plastic lawns and landscape lighting. The proposed approach offers valuable insights for precision forestry, ecological monitoring, and intelligent UAV-based remote sensing applications.

This study compares visualization types (3D, 2D, Oblique 3D, and Combined 2D+3D, coupled in a pairwise fashion for different tasks and scenarios), investigates their influence on decision-making across selected urban planning tasks (Site selection, Scenario Selection), and Distance Estimation as a baseline task that we assumed is relevant in both. Our goal was to inform the participatory urban planning process. In a controlled user study with 40 participants, we evaluated whether visualization type affects decision outcomes and distance estimation, complemented by participants’ visualization preferences before and after the experiment. The results confirm the previously well documented evidence that participants are considerably more successful with distance estimation with 2D visualizations, and their decisions vary depending on the examined visualization and task types. We observe that different formats support different task requirements, i.e., each visualization type exhibits distinct strengths depending on the task. These findings indicate that visualization choice in urban planning should be adapted to the task and context rather than treated as an interchangeable artifact.

With the rising number and severity of WUI wildfire episodes and the necessity to improve community preparedness, planning strategies have to be devised that integrate foresight into wildfires, with active community participation. This paper presents an intelligent collaborative environment that seeks to engage citizens, planners, and emergency services in co-creation of fire-resilient strategies through Agentic AI-driven wildfire simulations and mixed-reality visualization. A serious game environment is designed for hands-on exploration of alternative wildfire spread scenarios and community-scale prevention practices such as prescribed burning, fuel treatment through vegetation control, and structural hardening measures. The objective is to promote public awareness and adaptive behavior as well as provide science-based operational decision support to emergency responders in evaluating tactical options specific to terrain, infrastructure, and fuel conditions of the locale. The system operates on a Large Language Model (LLM)–powered agentic AI architecture designed to automate and orchestrate 2D and 3D wildfire simulations, providing guidance that supports users from diverse technical backgrounds. To give the results of the simulations, 3D web visualization and immersive holographic display were used to enable cycles of iterated explorations into fire spread in dense urban settings. With AI-assisted wildfire intelligence, this particular flow works through a set of intuitive interaction mechanics so that communities can evaluate risk levels, weigh alternatives for mitigation, and better prepare for an actual fire event.

Accurate spatial alignment is a key requirement for situated Augmented Reality (AR) in urban planning, where citizens and planners can visualize proposed designs in real outdoor environments. However, existing AR localization approaches often rely on smartphone GNSS, vendor-specific cloud anchors, or cloud-based visual positioning, which introduce accuracy limitations, privacy concerns, or dependencies that restrict their use in participatory planning workflows. This paper presents a privacy-aware on-device localization pipeline for outdoor urban planning scenarios. The approach aligns LiDAR scans captured on smartphones with pre-scanned reference point cloud tiles to enable stable and accurate placement of urban planning models. Approximate GNSS is used only to retrieve a relevant reference tile, while all preprocessing and registration steps are performed locally on the device. The pipeline combines voxel downsampling, local geometric descriptors, and global registration to estimate alignment without relying on GNSS for pose estimation or on cloud-based visual localization services. A mobile demonstrator was developed to support situated AR in urban planning scenarios, allowing users to explore design proposals directly in context. Initial validation under controlled conditions showed that the system can recover translations and rotations with errors on the order of a few centimeters, while processing times remained suitable for mobile use. The approach was also deployed in an urban planning case study and enabled stable outdoor visualization of planning elements on-site.

We present a new method for assessing visual walkability using 360° videos and an eye-tracking in Cinematic Virtual Reality (CVR). Visual walkability refers to the walkability perceived by pedestrians through visual stimuli in the urban environment. Our method uses semantic segmentation, viewport exposure, gaze measures, and a custom walkability questionnaire, enabling comparison between scene content, participant's viewport, and their gaze focus. The 10 videos used, including 2 calibration videos, exhibit distinct semantic characteristics, validated by segmentation analysis. Analysis of the 35 participants’ responses shows that walkability ratings at the video level correlate with several environmental parameters (e.g., road, sidewalk, sky) consistent with previous studies. However, these parameters do not have a similar influence in gaze-based visual attention analysis within the CVR setting, suggesting that CVR attention would requiere further work. Furthermore, our results suggest that unexpected semantic classes may also play a role in perceived walkability and should be considered exploratory pending further validation. This paves the way for further research on using CVR as an assessment tool for visual walkability and for developing methodological guidance on which visual cues are robust across measures (content/viewport).

Oceanographic data and their related simulation have a key role in addressing EU and UN societal challenges in marine environments. Visualising marine data is challenging for different visual-communication intents and audiences, despite existing guidelines on the subject. A main visual-design limitation for existing techniques is the co-visualization of multiple hydrodynamic field attributes in an accessible, comprehensible and engaging manner. This paper addresses this limitation in two ways: first, existing techniques for cartographic-oriented design of waterlines are adopted and extended towards multivariate hydrodynamic field datasets. Secondly, experimental results on the intermixing different visual-channel mapping of hydrodynamic attribute data are presented in a case study on ocean-flow patterns around the Hebrides island chain (UK). The results demonstrate a simultaneous co-visualization of up to five unique, independent scalar attributes in a comprehensible manner while preserving the geographic context. Moreover, best-practice guidelines are stated in conclusion of the experimental case study to help oceanographic practitioners adopt the presented technology in their professional workflows.

Artificial light at night (ALAN) degrades nocturnal ecosystems and complicates astronomical observation. Although all-sky imaging and GIS-based light-pollution mapping are well established in the analysis of light pollution, identifying local contributors to ALAN still requires time-consuming cross-comparisons, done in separate views, making light halo--source attribution slow and manual. We present an interactive system that addresses this gap by co-registering Sky Quality Camera all-sky imagery and OSM-derived candidate emitters (e.g., settlements, roads, aerodromes, industrial sites) in one observer-centered scene. The viewer is placed at the locations of the captured all-sky images in 3D digital terrain model-based scenes, realistically illuminated by the sky under selected conditions for an immersive view of nighttime scenarios. OpenStreetMap features are projected onto a surrounding sphere via inverse stereographic projection, with point markers and horizontal-extent indicators to support rapid visual matching between observed halos and plausible sources. Users can switch scenes and processed sky images, adjust projection parameters, and inspect scenes in VR or in an additional cylindrical projection for a panoramic desktop view. A companion web tool configures location classes and display ranges. The presented system primarily targets exploratory analysis, with its main contribution being the novel co-visualization of light sources and light halos; expert interviews positively validated this analytical focus. As a secondary outcome, the system's immersive first-person representation may also enrich educational communication and outreach on ALAN impacts.

Accurate and intuitive visualization of urban development projects is a persistent challenge in spatial planning and public participation. Recent advances in Extended Reality (XR) offer new opportunities to integrate geospatial data directly within the user’s real environment. This paper introduces GeoWebXR, an extension of the WebXR API designed to provide absolute georeferencing of the XR reference space via a standardized geopose. We present an outdoor proof-of-concept implementation that integrates a dual-antenna RTK GNSS receiver mounted on an XR headset. High-precision GNSS measurements are fused with the device’s local pose estimates to compute a consistent and accurate geopose, enabling decimeter-level alignment between virtual and physical environments.

Leveraging GeoWebXR, WebGL applications can render georeferenced 3D content in situ through a web browser. We demonstrate this capability using the iTowns geospatial visualization framework to deliver an XR experience for urban planning. The system supports both 1:1-scale in-situ visualization and reduced-scale overview modes, enabling seamless multiscale exploration of planning scenarios. To mitigate cognitive overload in dense urban contexts, we implement and evaluate several visualization and interaction strategies. We assess the usability and spatial appropriation enabled by our system, and discuss how it may support both expert analysis and citizen participation in urban planning processes.

LunarDEM2025 is a lunar topography model (±60°) created by fusing JAXA’s SLDEM2013, CAS’s CE2TMap2015 and NASA’s LOLA laser altimetry tracks. A tile-based, terrain-aware co-registration aligns photogrammetric DEMs to LOLA points, while a slope-constrained residual-compensation filter eliminates striping, voids and artefacts. The resulting dataset shows visibly smoother relief, smaller vertical biases and fewer tile-boundary discontinuities than its predecessor SLDEM2015. The product is ready for landing-site analysis, rover path planning and various other applications.

After completing the 1:2,500,000-scale (1:2.5 M) lunar geologic atlas, our team began exploring the techniques and work flows for compiling larger scale lunar geological maps. Geologic maps integrate multidimensional information such as lithology, structure, and geologic age. Using the Copernicus crater region (0°–16°N, 30°W–10°W) as a case study, this research develops a 1:1,000,000-scale (1:1 M) regional geologic map and, in turn, explores the lithologic and structural classification systems applicable to lunar geologic maps at different scales. Based on imagery, topography, spectral, gravity, and sample data, we analyze geologic features including impact craters, impact basins, compositions, and structures, and subsequently delineate geological units. In the study area, the Copernicus crater and Imbrium basin represent the most prominent geological events and can serve as benchmarks for relative age determination. The cross-cutting relationships among geological units, together with existing absolute age constraints (from isotopic dating and crater size-frequency distribution chronology), are used to establish the stratigraphic relationships among mapped features and layers, ultimately producing a regional geologic map. Based on this map, the geological evolution history of the region is reconstructed.

High-accuracy planetary mapping products are increasingly required for landing-site assessment, precision navigation, and future surface operations on the Moon and Mars. Although massive orbital remote sensing images are available, the geometric accuracy and spatial resolution of many existing mapping products is still insufficient for engineering applications. A major bottleneck is large-scale bundle adjustment, whose reliability is strongly affected by data quality, control network strength, as well as engineering experience. Compared with Earth observation photogrammetry, planetary mapping faces great challenges such as heterogeneous sensor models, complex illumination, sparse absolute control. This paper summarizes a practical quality control framework for large-scale bundle adjustment of planetary remote sensing images. The workflow is divided into four coupled stages: data preprocessing, control network construction, parameter setting, and accuracy evaluation. The framework is distilled from previous planetary mapping studies, open-source software platforms and our practical experience in processing tens of thousands of planetary images. Experiments using LRO NAC datasets demonstrate that satisfactory bundle adjustment results can be achieved when the proposed strategy is applied. The framework improves the overall efficiency, controllability, and reliability of large-scale planetary photogrammetric processing.

China has successfully carried out the Chang'e-1 to Chang'e-6 lunar missions and the Tianwen-1 Mars mission. In these missions, planetary photogrammetry and remote sensing technologies provide timely spatio-temporal information services across all phases of the missions, playing a crucial supporting role in ensuring the mission safety and scientific output. In the current era of artificial intelligence (AI), the deep integration of photogrammetry and remote sensing, geomatics, and artificial intelligence is gradually evolving into Spatio-Temporal Intelligence (STI). This paper presents an overview of the advances and applications of STI in China’s lunar and Mars explorations, and discuss the future directions of STI in deep space exploration.

Based on microwave radiometer (MRM) data from China's Chang'e (CE)-1/2 satellites, the Brightness Temperature Difference (TBD) technique offers a method for probing lunar regolith properties. However, its global application is compromised by systematic latitudinal biases and an unverified link to subsurface deposits.

This study introduces a novel parameter, the effective TBD (TBDeff), to overcome these limitations. The methodology first defines an equivalent TBD (eTBD), simulating the TBD for a location as if it were on the lunar equator to mitigate latitudinal effects. Recognizing inherent limitations in this simulation, a supplementary parameter (sup_TBD) is derived. TBDeff is then developed by integrating sup_TBD with the observed TBD (TBDobs) from CE-2 data.

Results demonstrate that TBDeff successfully removes latitudinal bias on a global scale, enabling clearer discrimination between lunar maria and highlands. Furthermore, extensive low-TBDeff signals in polar regions (>85°) suggest a new potential for detecting subsurface deposits in permanently shadowed areas.

Crucially, correlation analysis with (FeO+TiO₂) abundance reveals that TBDeff exhibits a significantly stronger relationship with regolith composition than traditional TBD or simple brightness temperatures (TB), especially at lower frequencies (reaching a correlation coefficient of 0.86 at 3.0 GHz). This confirms that (FeO+TiO₂) abundance is a key factor influencing the dielectric properties of subsurface materials, a effect previously obscured by latitudinal interference. The TBDeff method thus provides a more reliable tool for interpreting lunar composition from microwave data.

This study analyzed and compared in situ spectral obtained by the Chang’E-3(CE-3) and Chang’E-4(CE-4) rovers, laboratory spectra of Chang’E-5(CE-5) soils and remote sensing spectra. The remote sensing spectra exhibit significantly darker and shallower absorption features than laboratory or in situ spectra, reflecting highly weathered nature of the undisturbed lunar surface. The spectral upturn even just right >2 μm can be contributed by thermal emission, revealing micro-scale temperature variations and low thermal inertia of lunar soils. CE-5 sample spectra show significantly higher reflectance and absorption depths than in situ and remote sensing, indicating samples are fresh and couldn’t represent pristine/true lunar surface. The CE-5 samples provide a new ground truth for estimating the TiO2 content of young basalts, which have the largest uncertainty in TiO2 content. Contrary to traditional opinion, CE-3 in situ spectra revealed that the effect on the spectral slope caused by space weathering is wavelength-dependent: the visible slope (VS) decreases not increases. The optical effects of space weathering and TiO2 are identical: both reduce albedo and blue the spectra. This suggests that a new TiO2 abundance algorithm is needed.

The Virtual Campus Project at the University of Cologne (UoC) has as a main objective the creation of a highly detailed 3D model of the university campus and its publication and distribution through OGC 3D Tiles. Further objectives include the development of integrated applications leveraging this 3D model, such as a web-based 3D viewer, game engine-driven geospatial augmented reality (GeoAR) and virtual reality (VR) experiences, and an indoor positioning system utilizing 3D building models indoor geometries. This paper focuses on and details the methodology for developing and implementing the georeferenced 3D model and establishes an Open Geospatial Consortium (OGC) 3D Tiles-compliant Spatial Data Infrastructure (SDI). The main result is a Tool Suite or Software Framework and the description of the tool pipeline or workflows for collecting, creating and modelling the 3D geospatial data and publishing it as OGC 3D Tiles data. This framework ensures campus-wide 3D data accessibility through 3D Tiles standard clients, including Desktop GIS like ArcGIS or QGIS, game engines like Unity, Unreal or O3DE and Webmapping libraries like MapLibre, three.js or CesiumJS.

The assessment of sidewalk infrastructure for accessibility compliance is an important task in urban planning; however, traditional methods are often manual, subjective, and resource-intensive. This paper introduces BirdCV-LiDAR, a multimodal data fusion framework for an automated assessment of sidewalk infrastructure. The proposed approach integrates high-resolution bird's-eye-view (HR-BEV) imagery with aerial LiDAR point cloud (ALPC) data to automatically detect, measure, and assess sidewalk features for compliance with accessibility standards. By combining YOLO-oriented bounding box (OBB) models with precise LiDAR-based elevation data, the framework enables accurate dimensional and slope evaluations of sidewalk features, such as crosswalks and truncated domes. Validation with a 12-inch inclinometer shows that LiDAR-based slope measurements achieve 84.7% accuracy, with a root-mean-square error (RMSE) of 0.1152 meters for crosswalk width measurements. The framework achieves 81.0% accuracy in determining ADA-PROWAG compliance, providing an adaptable, expandable solution for improved urban accessibility assessments.

This paper proposes a micro-scale walkability metric based on harmonised indicators that supports pedestrian route planning, which prioritises pleasant environments alongside distance efficiency. The employed method quantifies street segments and crossings using geospatial indicators, including pavement width, slope, shade, adjacency to traffic, park context, and crossing type and width. Indicator values are transformed to percentile ranks to harmonise heterogeneous inputs and are aggregated into a single edge-level walkability score on a 0 to 1 scale. The score is integrated into a routing cost function that reduces edge cost with higher walkability, which favours calmer, greener, and wider links while bounding detours relative to the shortest path. The method also accommodates the incorporation of street-level perceptions through a structured survey instrument and a confidence-weighted fusion scheme.

The results show various spatial patterns. Central areas and park-adjacent segments exhibit higher scores, while steep, narrow, and traffic-exposed links score lower, and several suburban and foothill districts display reduced walkability. The comparison with a distance-only baseline shows selection of quieter alignments with modest length increases, indicating potential gains in perceived pleasantness.

Urban areas are continuously expanding outward with economic development and demographic growth, while simultaneously growing higher vertically. However, few efforts have been made to evaluate the impact of development priorities in different regions on urban sustainability, which limited our understanding of how urbanization has been affected by imbalanced evolution rhythm. Here we developed a 3D structure-based approach to assess volumetric urban expansion, as well as a refined evaluation system for assessing urban imbalanced growth trends. Results show that the 3D expansion patterns of urban areas exhibited significant heterogeneity globally. As urbanization accelerates, urban areas in the Global South are showing a trend of faster expansion accompanied by faster vertical growth. In addition, imbalanced growth types across different dimensions are significantly more complicated in the Global South than in the Global North, indicating the variance of development priorities is greater in the Global South. Furthermore, the imbalances are intensifying over time, as indicated by the temporal indices. Our study enhances the understanding of urban 3D patterns and imbalanced urban evolution, providing crucial insights for more balanced urbanization especially in the Global South.

Urban green spaces are important for residents’ physical and mental health, but green visibility is difficult to quantify in high-rise, high-density cities, especially across different height levels. To address this problem, this study proposes a stratified green visibility framework based on airborne LiDAR point clouds and a voxel model. Using the Dutch AHN5 dataset, the study area was converted into a unified 3D voxel space and classified into trees, grass, buildings, ground, and empty space. A voxel-level penetration probability model based on the Beer–Lambert law was introduced to represent the semi-transparent blocking effect of tree canopies, improving upon conventional binary visibility models. Multi-directional line-of-sight (LOS) tracing was then applied to calculate green visibility (GVI) and spatial openness (SOP) at different height layers. The results show that GVI is generally high around parks, large green spaces, and some enclosed courtyards, but its contribution from street trees is limited. Vertically, GVI decreases with height, while SOP tends to increase. Combining the two indicators helps identify different spatial types with distinct visual characteristics. The study demonstrates that airborne LiDAR, combined with voxelization and probabilistic 3D simulation, can effectively capture the vertical variation of urban GVI and support large-area assessment in high-density residential environments.

Accurate building-level mobile device positioning is critical for fine-grained location-based services and human activity analysis, as people spend 80–90% of time indoors. Existing techniques rely on dedicated infrastructure or dense fingerprinting, limiting scalability. This study proposes a lightweight, infrastructure-free framework integrating two core modules: 1) Indoor/outdoor classification via a random forest model trained on a multi-scene sample library, using satellite, Wi-Fi, Bluetooth, and cellular sensor features with similarity-guided training selection; 2) Building matching through a Bayesian inference model leveraging three-scale spatial features (device, building, area) and prior knowledge from anonymous crowdsourced data. Validated in Beijing, Nanjing, and Xi’an, the framework achieves over 90% overall precision for indoor/outdoor classification and ≥70% precision for building matching with satellite or Wi-Fi features alone. It requires no extra infrastructure or extensive labeled data, offering a scalable solution for smart city applications like population analytics, emergency response, and context-aware services across heterogeneous urban regions.

In Japan, the combined effects of declining birth and marriage rates have accelerated population decline, leading to spatial porosity and fragmentation in urbanised areas: a phenomenon known as “Urban spongification”. This study analyses local population distributions in order to identify localised low-population areas embedded within densely populated urban environments, with the aim of understanding spatial porosity and fragmentation in Osaka Prefecture. A multi-scale spatial autocorrelation approach was applied to detect the spatial extent of localised low-population areas, and results were compared between 1995 and 2020. The analysis further examined how the formation and change of localised low-population areas differ across Use Districts and according to long-term land-use transition histories. The findings reveal pronounced spatial variability within districts that cannot be captured by conventional population density metrics alone. The study demonstrates that the emergence, persistence, and transformation of localised low-population areas are closely related to zoning regulations and historical land-use processes. These results provide insights into the spatial processes contributing to urban porosity and fragmentation and offer a basis for future evaluations of residential inducement areas designated under Location Optimisation Plans.

Urban localization for autonomous driving requires accurate 6-DoF vehicle pose despite GNSS multipath, occlusions, and rapidly changing visibility. We fuse LiDAR, IMU, and GNSS in an error-state Kalman filter against a high-resolution (HR) map, aiming (i) to reduce LiDAR load without degrading accuracy and (ii) to improve robustness in building-sparse areas such as open junctions. The reference trajectory and HR map stem from a dedicated urban measurement campaign; Monte-Carlo simulations use ray-cast LiDAR, synthesized IMU, and GNSS tied to this trajectory so that only sensor noise is varied. A geometry-aware farthest-point sampling scheme prioritizes points informative for building/ground planes and pole-like structures, while an extended functional model introduces poles as additional vertical constraints. A retained-point rate of 10 % preserves trajectory-wide millimetrelevel and sub-milliradian accuracy, meeting in theory automotive requirements. Filter runtime is reduced by about 82 % relative

to the full LiDAR data. Compared with plane-only variants, the planes+poles configuration yields statistically significant but globally modest improvements in longitudinal, lateral, and yaw accuracy. More importantly, a sliding-window analysis reveals that it markedly stabilizes pose in plane-sparse junctions. Overall, the results suggest that task-aware subsampling preserves trajectory-wide performance while pole constraints add local robustness in challenging urban scenes; validation with real sensor logs remains necessary to confirm these accuracy margins, but the proposed filtering scheme shows promising potential for practical deployment.

Sensor data are increasingly used for monitoring and observation of spatiotemporal phenomena for diverse applications such as in flood management, urban traffic, air quality control, forest fire management, etc. Real time modelling and representation of such evolving phenomena is fundamental for efficient and timeliness decision-making processes. In the context of multisensory systems, where two phenomena (e.g.: air pollution index and windy condition) can both be sensed by networked sensors, analysing the relationship that hold between them is a major issue for decision making. Knowing if the pollution extent is expanding or contracting around a given spot or if it is within a windy zone can help in adopting more appropriate strategies. Sensing system equipped with rule-based reasoning engine to infer on spatiotemporal changes or topological relationship that holds between sensed phenomena with broad boundaries over time will provide decision-maker with adequate and non-ambiguous information. In this paper spatial changes and topological relationship about fuzzy-crisp object modelling the geometry of vague shape phenomena are conceptualized using an Extended Fuzzy Spatiotemporal Change Pattern (FESTCP) and a 5x5 Intersection model (I5x5M) respectively. The rule-based reasoning engine proposed in this paper is based on this conceptualisation. To evaluate our method, a simulated case study of air pollution in Quebec City is carried out. The results reveal that the proposed method captures well the spatiotemporal evolution of a given air pollution episode that served for an on-the-fly decision-making process in real life situations.

Reliable lane-level localization remains difficult for autonomous vehicles (AVs) when Global Navigation Satellite System (GNSS) observations are degraded by blockage, multipath, and non-line-of-sight reception in urban environments. This paper presents PointLoc, an Inertial Navigation System (INS)-centric locator aided by GNSS, monocular visual-inertial odometry (VIO), and High-Definition (HD) Vector Maps. The proposed method is formulated as an INS-centric error-state extended Kalman filter (EKF), in which the INS provides persistent state propagation, while GNSS, VIO, and map matching are incorporated as aiding updates according to their availability and reliability. This design preserves a unified position, velocity, and attitude solution and enables graceful degradation when some aiding sources become unavailable. The method is validated through real-vehicle experiments in Taichung Shuinan and Tainan Shalun under mixed GNSS conditions. The results show that PointLoc achieves the best overall full-route performance in Taichung Shuinan and remains broadly comparable to GNSS/INS/VIO, while still outperforming GNSS/INS, in Tainan Shalun. In the mapped GNSS-denied segment of Taichung Shuinan, PointLoc effectively suppresses vertical drift and substantially improves three-dimensional positioning. The mapped-road analysis further shows that the INS-centric design avoids the planar instability observed in a vision-centric benchmark and provides a more continuous localization solution.

Conventional laser scanning techniques (such as in a Terrestrial Laser Scanner or Mobile mapping), whether used in a static or mobile mode require the object of interest to remain stationary during the scanning stage. Any motion of the object during scanning results in the apparent distortions in the resulting point cloud. The authors in Goel and Lohani (2014b) proposed a motion correction technique to estimate the 3D geometry of a moving object, utilizing a fusion of inertial and GNSS (Global Navigation Satellite Systems) sensors and transformation of the resulting point cloud to an object body coordinate system (OBCS). This paper aims to carry out the experimental validation and performance analysis of the motion correction method. Field experiments are designed and conducted in three phases to verify the correctness of the method. Through this, the paper aims to uncover insights into the performance of the motion correction algorithm and provide the first experimental validation of the proposed technique.

Wearable sensors are essential for gait analysis outside of traditional laboratory environments. However, selection of the right sensor technology involves several trade-offs. Inertial Measurement Units (IMUs) offer high temporal resolution which are ideal for detecting gait events but they suffer from drift. Ultra-Wideband (UWB) provides stable spatial data, but are less precise for detecting event timing. This paper presents a comparative study of three distinct foot-mounted sensor methodologies for heel strike detection and cadence estimation: (1) IMU-Only approach, (2) UWB-Only approach, and (3) a multi-sensor IMU+UWB fusion approach. Each method is evaluated against a camera-based ground truth system using data from four subjects. Results show the IMU-Only method is inconsistent, with moderate event precision (Avg. F1: 0.798), temporal accuracy (Avg. MAE: 47.99 ms), and subject-dependent cadence accuracy (Avg. Acc: 89.59%). The UWB-Only method provides robust event detection (Avg. F1: 0.811) with similar temporal error (Avg. MAE: 49.0 ms) but is exceptionally accurate for cadence estimation (Avg. Acc: 96.94%). The IMU+UWB fusion approach achieves the highest event precision (Avg. Precision: 0.835) and the best temporal accuracy (Avg. MAE: 46.51 ms), while also maintaining robust cadence accuracy (Avg. Acc: 95.62%). In conclusion, while the UWB-Only method is ideal for cadence-only applications, the IMU+UWB fusion approach provides the best overall balance of high event precision, superior temporal accuracy, and reliable cadence estimation.

Accurately registering building footprints from heterogeneous datasets with LiDAR data remains a critical challenge in urban mapping and 3D reconstruction. The objective of this work is to register source data, defined as 2D cadastral vector footprints from structured, regularized, or manually-verified datasets to target building footprints derived from classified aerial LiDAR. LiDAR provides direct 3D information with precise footprint positioning and high spatial resolution, enabling a geometrically reliable representation of dense 3D structures. Conversely, source datasets are not always up-to-date, and may exhibit geometric distortions such as translational offsets, rotational deviations, or local deformations, yet they remain valuable due to their structured organization and metadata content. To enhance geometric fidelity while preserving semantic structure, we propose a practical framework for non-rigid polygon registration that adjusts the geometry of cadastral footprints toward LiDAR-derived targets. The framework consists of two core components: (1) establishing correspondences between source and target polygons, and (2) minimizing a robust distance function that governs the registration process. Three deformation models are introduced: a rigid model allowing translations only, a semi-rigid model allowing deformations while keeping the overall structure of source footprints, and a non-rigid model allowing rotations. We evaluate our method by aligning real cadastral datasets to aerial LiDAR data. The results confirm the effectiveness and robustness of the proposed framework in the context of 2D polygonal cadastral data. This work thus represents the first practical solution for non-rigid polygon registration in this domain.

This paper presents a mask-aware multi-stage depth enhancement framework for digital documentation of historical timber surfaces using RGB–Stereo-IR fusion. Accurate geometric recording of aged wood features such as wooden knots remains challenging due to uneven illumination and weak texture. The proposed pipeline, which aims to stabilise depth geometry under uneven illumination and low-texture surface conditions, integrates object detection, instance segmentation and confidence-guided depth refinement across three stages: (A) TV(total variation)-regularized mask-aware refinement, (B) confidence-weighted multi-view fusion, and (C) patch-based stereo reconstruction. Experiments on historical timber beams under varying illumination demonstrate improved depth completeness and geometric consistency, achieving a residual standard deviation below 0.6 mm in bright scenes and stable reconstruction in low-light conditions. The framework offers a practical solution for depth reconstruction of cultural heritage timber, supporting more reliable feature detection and analysis.

Large-scale 3D reconstruction has emerged as a key research in the fields of photogrammetry and computer vision. 3D Gaussian Splatting (3DGS) has become a mainstream approach due to its efficient rendering, but it confronts critical challenges in large-scale scenarios: excessive memory overhead and inadequate geometric accuracy. Meanwhile, the traditional Structure from Motion and Multi-view Stereo (SfM-MVS) framework, despite its cumbersome process, continues to exhibit robust performance. Notably, a systematic evaluation comparing these two paradigms in large-scale scenes remains absent. To address this, we develop a unified verification framework to evaluate the texture rendering quality and geometric reconstruction precision of several recent methods using real-world datasets. The results indicate that SfM-MVS methods still maintain an advantage in the completeness and accuracy of geometric reconstruction. In contrast, 3DGS methods have achieved breakthroughs in local accuracy or rendering-geometry synergy, yet their global consistency requires further improvement.

3DGS-based methods often render transient distractors in 3D scenes as significant floating artifacts. Existing works for removing transient distractors suffer from under-identification or over-identification, resulting in residual transient distractors affecting reconstruction quality or loss of scene information, preventing the reconstruction of fine details. To address these challenges, we propose RobustGauss. We first rely solely on the cosine similarity of DINOv2 features to robustly predict uncertainty masks and accurately identify the main regions of transient disturbances and their corresponding shadows. Due to the limited resolution of DINOv2 features, we use high-resolution image residuals to refine the edges of the initial uncertainty masks, thereby accurately identifying all transient distractors and minimizing their impact on 3D scene reconstruction. Experiments on two challenging datasets demonstrate that our method achieves state-of-the-art performance.

N/A

We present EMVSNet, a sampling-free Multi-View Stereo (MVS) method that, to the best of our knowledge, is the first to integrate Evidential Deep Learning into MVS. Given a set of overlapping images, our method predicts a depth value together with its associated uncertainty per pixel of a reference image, incorporating uncertainty from aleatoric and epistemic sources. Specifically, we use an existing convolutional neural network architecture designed for MVS as backbone and extend it to regress evidential parameters per pixel, describing the probability distribution over the depth corresponding to this pixel. In contrast to existing MVS methods that often neglect epistemic uncertainty or obtain it via sampling at inference, our evidential formulation does not require sampling, but enables single-pass inference. We evaluate the uncertainty estimation capabilities of our method using two publicly available datasets and compare the depth predictions against a deterministic variant. The experimental results demonstrate that EMVSNet achieves competitive depth accuracy while, at the same time, providing uncertainty estimates that enable us to reliably rank depth estimates according to their risk of being incorrect and to automatically identify out of distribution data. Our model shows only slightly increased inference time compared to a deterministic baseline while giving comparable uncertainty estimates to an computationally expensive sampling based approach, marking a first step towards real-time capable uncertainty estimation for image-based 3D reconstruction.

Object completion networks typically produce static Signed Distance Fields (SDFs) that faithfully reconstruct geometry but cannot be rescaled or deformed without introducing structural distortions. This limitation restricts their use in applications requiring flexible object manipulation, such as indoor redesign, simulation, and digital content creation. We introduce a part-aware scaling framework that transforms these static completed SDFs into editable, structurally coherent objects. Starting from SDFs and Texture Fields generated by state-of-the-art completion models, our method performs automatic part segmentation, defines user-controlled scaling zones, and applies smooth interpolation of SDFs, color, and part indices to enable proportional and artifact-free deformation. We further incorporate a repetition-based strategy to handle large-scale deformations while preserving repeating geometric patterns. Experiments on Matterport3D and ShapeNet objects show that our method overcomes the inherent rigidity of completed SDFs and is visually more appealing than global and naive selective scaling, particularly for complex shapes and repetitive structures.

Panoptic segmentation requires the simultaneous recognition of countable thing instances and amorphous stuff regions, placing joint demands on long-range context modelling, multi-scale feature representation, and efficient dense prediction. Existing convolutional and transformer-based methods struggle to satisfy all three requirements concurrently: convolutional architectures are limited in their capacity to model long-range dependencies, while transformer-based methods incur quadratic computational cost that is prohibitive at high resolutions. In this paper, we propose MambaPanoptic, a fully Mamba-based panoptic segmentation framework that addresses these limitations through two principal contributions. First, we introduce MambaFPN, a top-down feature pyramid that leverages Mamba blocks to generate globally coherent, multi-scale feature representations with linear computational complexity. Second, we adopt a PanopticFCN-style kernel generator that produces unified thing and stuff kernels for proposal-free panoptic prediction, enhanced by a QuadMamba-based feature refinement module applied at multiple network stages. Experiments on the Cityscapes and COCO panoptic segmentation benchmarks demonstrate that MambaPanoptic consistently outperforms PanopticDeepLab and PanopticFCN under comparable model sizes, and matches or surpasses Mask2Former on Cityscapes in PQ and AP while requiring fewer parameters.

Single-view 3D reconstruction from monocular aerial imagery presents a fundamental challenge in remote sensing due to the inherent scale ambiguity and the complex geometry of urban environments. Traditional regression-based methods often struggle to recover high-frequency structural details, leading to over-smoothed or noisy outputs. To address this, we introduce GeoPrior-Diff, a novel two-stage framework that leverages the generative capabilities of Latent Diffusion Models to reconstruct high-fidelity 3D point clouds.

Unlike direct generation approaches, our method explicitly bridges the domain gap between 2D texture and 3D structure by utilizing an intermediate geometric prior. In the first stage, we predict an oblique normal map from the input satellite imagery, capturing essential surface orientation and structural boundaries. In the second stage, this normal map serves as a strong conditioning signal for a probabilistic diffusion model, guiding the denoising process to synthesize accurate 3D point clouds. Preliminary results demonstrate that decoupling geometric estimation from point generation significantly enhances structural consistency and reduces artifacts compared to baseline methods. This work highlights the potential of using generative priors for robust 3D urban modeling from limited data.

This presentation examines the integration of Earth Observation (EO) data into Canada’s impact assessment (IA) processes, highlighting the progress and applications of the Earth Observation for Cumulative Effects – Phase 2 (EO4CE-2) program. Despite rapid growth in EO data acquisition, analytics, and delivery systems, uptake by IA practitioners has been limited due to persistent barriers such as awareness of EO capabilities, data accessibility, and analytical capacity. EO4CE-2, led by Natural Resources Canada’s Canada Centre for Mapping and Earth Observation (CCMEO), aims to address these challenges by providing high-quality, standardized EO datasets and operational frameworks to support transparent, data-driven IA processes.

EO4CE-2 has produced a library of EO-derived products leveraging decadal satellite records and advanced machine learning, enabling comprehensive analysis of land use, water resources, vegetation, lake and river ice, and terrain stability. These datasets allow decision-makers to evaluate environmental status and trends.

A key application is the Regional Assessment of the Ring of Fire (ROF) area in northern Ontario, where eleven EO-based indicators—covering water systems, wildlife habitat, forest ecosystems, permafrost, and terrain deformation—support assessment priorities such as environmental health, social equity, and community well-being. Indigenous communities have played a central role in validating these indicators and contextualizing EO data. The results demonstrate that combining satellite observations with local knowledge enhances regional assessments, supports sustainable resource management, and informs evidence-based decision-making. This presentation highlights EO4CE-2’s achievements, challenges, and lessons learned in advancing the use of EO for cumulative effects assessment in Canada.

Forest biomass plays a central role in carbon accounting, climate modeling, and sustainable forest management. However, large-scale biomass estimation remains challenging due to the limited spatial coverage of field inventories and the inherent spectral saturation issues of optical remote sensing in dense forest canopies. This study presents an operational workflow for mapping above-ground biomass (AGB) across southern Québec using multi-source Earth observation data and machine learning implemented in Google Earth Engine. The approach integrates Sentinel-2 optical composites, Sentinel-1 dual-polarization SAR metrics, and a high-resolution 1-m canopy height model with detailed plot-level biomass derived from Québec’s Placettes-Échantillons Permanentes (PEP) network. A Gradient Tree Boosting model was trained on 4,083 quality-controlled field plots to capture species, structural, and spectral variability. Validation results show strong agreement between predicted and observed biomass (R² ≈ 0.76, RMSE ≈ 14.4 Mg ha⁻¹), demonstrating the value of fusing optical, radar, and structural predictors. The resulting biomass and carbon maps provide actionable information for forest monitoring, regional reporting, and environmental decision-making. This contribution highlights the effectiveness of cloud-based multi-sensor fusion for operational AGB estimation and offers a scalable methodology applicable to broader Canadian forest regions.

Seasonal snow remains a critical component of Canada’s water cycle, yet consistent, high-resolution monitoring of snow water equivalent (SWE) is still lacking at national and hemispheric scales. The Terrestrial Snow Mass Mission (TSMM) proposes a dedicated dual-frequency Ku-band radar satellite designed to deliver spatially continuous SWE estimates at 500 m resolution with a 5–7 day revisit rate. To prepare the scientific foundations of this mission, the TSMM Academic Consortium has expanded to 16 Canadian institutions and now integrates data from over 40 long-term snow research sites.

Between 2024 and 2026, the consortium conducted ten coordinated field campaigns across mountainous and Arctic environments, in collaboration with Environment and Climate Change Canada, the Canadian Space Agency, and the European Space Agency. These campaigns combined ground-based and airborne Ku-band radar, detailed snowpit measurements, microstructure characterization, UAV surveys, and GNSS mapping. Joint ESA–TSMM activities at Cambridge Bay further enhanced Ku-band validation in deep Arctic snow.

Recent advances include improved dual-frequency Ku-band inversion methods, refined radiative transfer models, enhanced wet/dry snow classification, and integration of radar-derived SWE into snow model simulations and CaLDAS assimilation frameworks. Together, these developments confirm TSMM’s feasibility and scientific readiness.

This contribution summarizes the consortium’s field results and retrieval advances, demonstrating the mission’s potential to provide operational SWE monitoring essential for hydrology, climate science, wildfire preparedness, and Arctic environmental security.

The disaster resilience in urban environments remains a critical yet often underexplored component of sustainable development, particularly in densely populated regions where schools and community shelters serve as vital emergency infrastructure. Despite their importance, the systematic assessment of these shelters’ suitability is frequently overlooked, leading to disparities in safety, accessibility, and preparedness during crisis events. This research introduces a comprehensive, data-driven framework for evaluating the suitability of educational institutions and community shelters using Gaussian Process Regression (GPR). The proposed model integrates multiple geospatial and infrastructural parameters, including environmental risk exposure, proximity to fault lines and water bodies, structural integrity, road connectivity, and population density. By modeling the complex nonlinear relationships among these significant factors, the Gaussian Process Regression (GPR)-based approach predicts shelter safety scores that reflect the relative resilience and accessibility of each location. The predicted scores are spatially visualized using interactive geospatial mapping tools, allowing decision makers to easily identify safer zones or shelters and high-risk clusters across Delhi. The areas with higher scores correspond to shelters with strong infrastructure and better access to emergency resources and open spaces, whereas lower-scoring regions indicate vulnerable areas in need of immediate policy attention and structural reinforcement. The outlier detection techniques further enhance the interpretability of results by identifying anomalous schools with unusually high or low suitability for deeper investigation. The model’s performance, evaluated through five-fold cross-validation, reveals variability in Mean Squared Error (MSE) across folds, indicating sensitivity to spatial heterogeneity and highlighting potential improvements through hyperparameter tuning and ensemble learning strategies.

Rapid urbanization has profoundly reshaped the spatial dynamics of population distribution, environmental quality, and resource utilization, particularly in megacities such as Beijing. As population density, industrial activity, and transportation intensity continue to rise, air pollution has become a key constraint on sustainable resource development and urban livability. In recent years, the integration of Earth Observation (EO) with geospatial analytics has provided new opportunities for monitoring, modeling, and managing urban environmental systems.

For instance, Liu et al. employed complex network theory to analyze regional air quality variations in the Yangtze River Delta[1], while Rabie et al. developed a CNN–Bi-LSTM hybrid framework to predict spatially resolved AQI in megacities[2]. Similarly, Similarly, Ma et al. used a temporal-encoded Informer model to forecast AQI in northern China[3],and Ahmed et al. demonstrated that EO-derived hydro-climatological variables can substantially enhance AQI prediction accuracy when combined with deep learning models[4].Moreover, Sarkar et al. proposed an effective hybrid deep learning model for AQI prediction, which further validates the potential of hybrid approaches in capturing complex urban air pollution patterns[5].However, most existing studies emphasize temporal forecasting or algorithmic improvement, while the spatial interaction between population distribution and air quality remains insufficiently explored. To bridge this gap, this study develops an EO-supported geospatial framework that integrates demographic and environmental data to analyze spatial heterogeneity and exposure inequality in Beijing, providing data-driven insights for sustainable resource and environmental governance.

In this study, we compared the performance of four machine learning (ML) models, including Multiple Linear Regression (MLR), Random Forest (RF), XGBoost, and Support Vector Regression (SVR), for predicting and mapping key water quality parameters, namely dissolved oxygen (DO), total phosphorus (TP), total nitrogen (TN), and turbidity in Fanshawe Lake using two distinct satellite datasets: the high-resolution PlanetScope and the Sentinel-2 imagery.

Eleven commonly used spectral indices sensitive to suspended particles and algae were derived from PS and Sentinel-2 imagery, combined with in situ measurements collected from 2018 and 2024 to train and validate the models. We evaluated the ML models using R², mean absolute error (MAE), and root mean squared error (RMSE). Our study shows that using machine learning with satellite imagery can provide encouraging predictions of key water quality indicators in Fanshawe Lake. There are certain benefits of using high spatial and temporal resolution PS satellite imagery instead of Sentinel-2 datasets to capture localized changes in water quality parameters. The Upper Thames River Conservation Authority can use these results to predict when algal blooms might occur in Fanshawe Lake. Future research may investigate the capture of seasonal trends through the integration of additional field and satellite datasets with time-series models, such as Long Short-Term Memory.

Climate change is rapidly affecting glaciers and seasonal snow in high-altitude regions, which in turn threatens water resources and mountain ecosystems. In this study, I aim to understand and quantify these climate-driven changes by combining data from Sentinel-1 radar and Sentinel-2 optical satellite imagery. By analyzing datasets collected over multiple years, I can observe how glaciers are retreating, snow cover is changing, snow grain size is evolving, and seasonal melt patterns are shifting. To achieve this, I use a combination of advanced spectral and radar indices along with machine learning techniques to extract detailed information about snow and glacier characteristics and track their changes over time. These results allow me to evaluate when snow melts and how it may affect downstream water flow, which is essential for sustainable water management and maintaining ecosystem health. I also make use of cloud-based platforms like Google Earth Engine to efficiently process large volumes of satellite data. By integrating AI-driven analysis with remote sensing, I can produce accurate, large-scale maps and insights that help predict future trends. The outcomes of my study are not only important for understanding how climate change is impacting glaciers and snow in my study area but also provide a framework that can be applied to other mountain regions around the world. Ultimately, my research offers valuable information for planning climate adaptation strategies and ensuring the resilience of both water resources and mountain ecosystems.

High-rise buildings are fundamental components of modern urban infrastructure, and their structural safety under dynamic loads such as wind and earthquakes is critically important In recent years, ground-based radar interferometry has been widely employed for monitoring vibrations and deformations in tall structures, owing to its high precision, non-contact operation, and full-field measurement capability. In practical monitoring, displacement signals are affected by various types of noise, leading to unstable and nonlinear variations in the signal. This makes it difficult to accurately obtain structural vibration characteristics (such as frequency and damping ratio) and micro-deformation data with precision. Conventional denoising techniques are often applied for noise reduction. Nevertheless, these methods exhibit notable limitations. Bandpass filtering requires a predefined frequency passband, becoming ineffective in cases of spectral overlap between signal and noise. Wavelet denoising lacks adaptability due to its strong dependence on the selected wavelet basis and decomposition level, often introducing signal distortion. Kalman filtering, meanwhile, relies on an accurate state-space model, the construction of which is challenging for complex high-rise structures, thereby limiting its practical utility. In response to these challenges, this paper proposes a fully adaptive denoising method based on CEEMDAN, incorporating dual criteria of energy distribution and correlation. The proposed approach effectively processes non-stationary and nonlinear signals while avoiding the subjectivity associated with basis selection in traditional methods. It significantly improves both mode separation accuracy and denoising reliability, establishing a robust foundation for structural state assessment and safety early warning based on radar monitoring data.

Localisation remains a crucial aspect of robotic design. It forms the basis of any kind of autonomous navigation for drones, cars and other specialized robots. This is usually achieved using a Simultaneous Localisation and Mapping (SLAM) algorithm, which uses an input sensor to localise the robot within a map that is created simultaneously. The input sensors are either cameras, which provide visual data, or Light Detection And Ranging (LiDAR) sensors, which automatically deliver a point cloud up to surveying quality. In recent years, LiDAR inertial odometry (LIO) algorithms, which combine measurements from a LiDAR sensor and inertial measurements from an IMU, have become more popular. These algorithms do not use a previously recorded map, but rather create their own map during runtime. This paper contributes an improvement to the precision by integrating a predefined 3D global point cloud map into the localisation algorithm. Over the course of multiple experiments in different testing scenarios, we have achieved a 71% reduction of the distance error for localisation, while there was no significant change regarding the orientation error. This makes the presented system a suitable localisation option for real-world robotic operations at construction sites.

Reliable visual localization is essential for long-range VTOL UAV mapping in GNSS-degraded environments. This paper presents a quantitative evaluation framework for monocular ORB-SLAM3 using a 66.48 km multi-altitude UAV mission and aerial-triangulation-derived camera poses as reference data. The workflow associates SLAM and reference trajectories by image key, applies Sim(3)-based metric alignment, corrects coordinate-axis inconsistency, and refines attitude by a global rotation offset, enabling full-mission and segment-level comparison in a common metric frame. The evaluation covers four altitude segments, namely 100, 150, 200, and 250 m AGL, under three protocols: No-Loop (NL), With-Loop Global Slice (GS), and With-Loop Local Re-Sim(3) (LR). For the full mission, the proposed alignment achieves a 3D position RMSE of 7.41 m over 5330 matched frames and substantially reduces the geometric deformation observed in the S+T baseline. Segment-level results show a strong altitude dependency in the isolated NL runs, with 3D RMSE decreasing from 22.95 m at 100 m to 5.49 m at 250 m. Among the three protocols, LR consistently yields the best segment-level position accuracy, reaching 4.00, 8.26, 3.94, and 3.92 m at 100, 150, 200, and 250 m, respectively. Long-range analysis further shows that the trajectory remains globally bounded, while cumulative 3D endpoint drift increases from 0.35 m at 50 m to 10.66 m at 25.6 km. These results indicate that ORB-SLAM3 can support large-scale trajectory estimation for UAV mapping, but its evaluated quality depends strongly on alignment, segmentation, and evaluation strategy.

Tropospheric delays significantly hinder accurate InSAR deformation mapping, and their complex spatiotemporal vari-ability makes effective mitigation challenging. When GNSS are available, conventional functional models interpolate GNSS-derived delays to unobserved locations, but their low-order form mitigates only long-wavelength errors and neglects the stratified component. In phase-based correction, temporal low-pass filters such as the Gaussian filter suppress high-frequency turbulence but ignore the strong distance/elevation-dependence of tropospheric delays, making the results highly sensitive to the chosen time window .

In response, we adopt a Least Squares Collocation (LSC) scheme, an effective approach that treats spatially correlated turbulence as a stochastic variable, characterizes it through a variance–covariance model , and estimates it jointly with other deterministic parameters. With external GNSS data, a joint correction that accounts for both the strati-fied and turbulent components is constructed, and are simul-taneously estimated using LSC. For the data-driven case, the deformation phases are parameter-ized by a time-domain polynomial, while the turbulence are treated as spatially correlated stochastic variables defined by spatial variance-covariance functions. LSC is employed to estimate the deformation model parameters through sliding time-window filtering process .

We design a voice-interactive indoor positioning method that jointly utilizes spatial “near” relationships extracted from verbal descriptions and multiple sensor sources

Real-time dense SLAM must balance geometric fidelity and computational efficiency for autonomous navigation. Explicit mapping methods provide stable global structure and fast updates, but suffer from discretization artifacts and memory overhead. Implicit neural representations capture continuous surfaces and fine details, yet require expensive optimization and are sensitive to initialization. Existing hybrid approaches combine both paradigms, but often allocate neural refinement inefficiently and remain vulnerable to pose errors. To address these limitations, we propose a selective hybrid dense mapping framework that couples a scene-wide TSDF backbone with quality-guided implicit local refinement and residual-guided sliding-window pose feedback. Neural refinement is activated only in low-quality regions identified by multi-indicator assessment, while keyframe poses are re-optimized using residuals from explicit raycasting and implicit rendering. Experiments on TUM RGB-D and Replica demonstrate improved mapping accuracy, localization robustness, and real-time efficiency.

We present a simultaneous calibration of the 6 installation parameters (3 boresight angles and 3 lever arm offsets) for a mobile LiDAR system on a hydrographic platform using spherical targets. This algorithm finds the boresight angles and lever arm offsets that minimize the sum of positive distances from points to their corresponding sphere surfaces. The calibration method is first developed and tested using synthetic data generated by a ray-tracing algorithm using a line-sphere intersection model and is subsequently validated using real scan data. The spherical targets are installed on tripods at the calibration site, where their centers are surveyed using postprocessed GNSS observations. The RMS error for the distance between the surveyed sphere centers and the fitted sphere centers is 3.6 cm, which we attribute to the propagation of GNSS and LiDAR scanner errors. The MATLAB code developed for the simultaneous estimation of a complete set of 6 LiDAR installation parameters using spherical targets is available as open-source software on GitHub.

Efficient road surface condition evaluation is critical for ensuring transportation safety, maintaining ride quality, and supporting informed pavement management decisions. Conventional methods such as laser profilers and specialized inertial systems provide highly accurate measurements but are expensive, limited to specialized vehicles, and difficult to scale for network-level monitoring. To address these limitations, this study presents an accelerometer-based framework that leverages survey-, mapping-, and consumer-grade GNSS/INS units to detect pavement surface anomalies in a cost-effective and scalable manner.

Vertical acceleration data were collected using three inertial systems mounted on the Purdue Wheel-based Mobile Mapping Systems: the high-accuracy PWMMS-HA, the ultra-high-accuracy PWMMS-UHA, and a compact low-cost OpenIMU paired with a SparkFun GPS-RTK2 unit. All systems were driven along a 59 km closed-loop roadway network in West Lafayette, Indiana, capturing diverse pavement conditions under identical driving trajectories. The proposed pipeline includes two complementary anomaly detection approaches. The first applies an Isolation Forest model, an unsupervised machine learning technique that identifies abnormal vibration patterns using statistical window-based features. The second employs an Adaptive Threshold method that flags acceleration windows exceeding a dynamic statistical threshold. Both methods categorize detected anomalies into mild, moderate, and severe levels.

Across all IMU grades, the Isolation Forest detected 962–965 anomalies, while the Adaptive Threshold identified 989–996 anomalies, with more than 91% spatial agreement between sensors and over 96% consistency between detection approaches. Results demonstrate that even low-cost inertial sensors reliably capture pavement disturbances.

Accurate three-dimensional (3D) data acquisition of indoor environments remains a challenging and resource-intensive task, particularly for fully furnished spaces. This study presents the development and implementation of a low-cost wearable surveying system for efficient 3D indoor data acquisition. The proposed system integrates a Livox Mid-360 LiDAR sensor and an RGB camera mounted on a helmet, both controlled via a min-PC unit for synchronized data collection. The captured LiDAR frames and inertial measurement unit (IMU) data are fused with RGB imagery using a Simultaneous Localization and Mapping (SLAM) framework to generate 3D reconstruction of interior structures. The resulting point clouds and wall models are evaluated based on RANSAC line fitting method to assess their geometric accuracy and structural consistency. Moreover, a ground truth measurements were collected to verify the absolute accuracy of the resulting point clouds. The proposed approach demonstrates the potential of cost-effective, portable solutions for indoor 3D mapping and documentation with a cm-level of accuracy.

Hongtu-1 (HT-1) InSAR satellite, as an innovative multi-baseline X-band InSAR system, employs a four-satellite cartwheel formation to achieve single-pass interferometry. HT-1 has been extensively used for topographic mapping, disaster monitoring, and assessment. This research focuses on assessment of HT-1's capability for high-precision forest vertical structure inversion, especially for forest height estimation (X-band SAR is sensitive to canopy of forest due to the higher frequency). This paper develops a complete interferometric processing for HT-1 multi-baseline data. By applying two representative InSAR techniques to HT-1 multi-baseline InSAR data, this results demonstrate the forest vertical structure profile and canopy height map derived over the test site, which shows good agreement with the LiDAR data. The results confirm HT-1's feasibility for tomography and demonstrate the potential of multi-baseline satellites for future missions.

Hyperspectral images (HSIs) provide rich spatial–spectral information for applications such as environmental monitoring, land cover mapping, and mineral exploration. However, their practical utility is often severely degraded by mixed noise (Gaussian, impulse, and other unstructured components), striping artifacts, and partially missing data, especially in bands affected by strong water vapor absorption. Existing methods typically treat these degradations separately and struggle to jointly correct them within a unified framework.

This work presents an integrated HSI reconstruction method that couples a low-rank decomposition model with a hybrid supervised–unsupervised deep architecture. The HSI is factorized into spatial abundance maps and spectral endmember signatures, which are respectively modeled by a Transformer-based abundance reconstruction network and a 1D convolutional endmember smoothing network. The abundance network is first supervisedly pre-trained on large-scale natural image datasets and then fine-tuned, together with the spectral network, using unsupervised loss terms tailored to spatial and spectral fidelity. A weighted group-sparse regularization is further introduced to explicitly capture striping noise and constrain the learned subspaces.

Extensive experiments on simulated Washington DC Mall data and real Gaofen-5 (GF-5) satellite imagery demonstrate that the proposed method effectively suppresses unstructured noise, removes striping artifacts, and recovers missing information, achieving superior visual quality, higher spectral fidelity, and fewer artifacts compared with state-of-the-art baselines.

Static Terrestrial Laser Scanning (TLS) and SLAM-based Portable Mobile Laser Scanning (PMLS) are increasingly adopted in Cultural Heritage (CH) documentation, but their suitability for Historic/Heritage Building Information Modelling (HBIM) depends on both data quality and acquisition conditions. This paper compares a 2022 TLS survey and a 2025 handheld PMLS survey of San Giacomo Church (Como, Italy) to assess whether the latter can reliably support HBIM-oriented documentation. The methodology combines dataset-level comparison and ROI-based analysis on four stable architectural elements: apsis, pillar, timber roof truss, and central dome. Three complementary metrics were used: a local density proxy, a scale-dependent coverage ratio, and M3C2 distance statistics for geometric agreement. Results show that PMLS is consistently less dense than TLS but remains effective for 1:100 scale documentation and, in several cases, also for 1:50. Statistics on M3C2 distance remain generally within centimetric ranges, indicating good local agreement where surfaces are effectively observed. The study demonstrates that sensor suitability not only depends on the geometric complexity but also on sensor-to-surface distance, visibility, and acquisition geometry, supporting hybrid TLS–PMLS workflows for CH HBIM.

This study examines the potential of RTK GNSS-integrated close-range photogrammetry for documenting cultural heritage without the need for ground control points (GCPs). The research focuses on evaluating the GEOSTIX-X5 GNSS receiver, which enables direct synchronisation with a camera via the flash hot shoe, providing precise time-stamping of image capture events. The case study was conducted at Tomb 8 of the Tombs of the Kings in Paphos, a UNESCO World Heritage Site, and compares two datasets: a conventional photogrammetric survey from 2022 using GCPs and a 2025 survey employing GNSS-assisted photogrammetry. Both terrestrial and UAV imagery were acquired and processed in Agisoft Metashape, with accuracy assessment performed through cloud-to-cloud comparison in CloudCompare. Results indicate that the GNSS-integrated approach achieved single centimetre-level accuracy and no systematic scale errors. The findings demonstrate that RTK GNSS-assisted photogrammetry can significantly reduce fieldwork complexity while maintaining high accuracy, offering a promising alternative for heritage documentation where GCP placement is impractical or undesirable.

UAS photogrammetry has become an efficient solution for acquiring high-resolution geospatial data for urban mapping, environmental monitoring, and 3D modelling. However, mission planning still involves a trade-off between data quality and operational efficiency, particularly regarding flight altitude, which directly affects ground sample distance (GSD), point cloud density, and positional accuracy. This study evaluates the influence of flight altitude through a controlled comparison of two urban photogrammetric surveys: a low-altitude flight at 61.2 m (GSD = 1.56 cm/pix, 420 images) and a higher-altitude flight at 121 m (GSD = 3.11 cm/pix, 116 images). Both surveys used RGB cameras with equivalent image resolution mounted on different platforms, which constitutes an experimental limitation, while overlap and processing parameters were kept constant. The results show that the lower-altitude flight produced denser data and better geometric performance, with lower reprojection error and lower check point RMSE. In contrast, the higher-altitude flight provided greater operational efficiency, covering a larger area with fewer images and lower computational demand. These findings indicate that both strategies are technically viable but suited to different objectives: lower altitudes favour geometric detail and positional accuracy, whereas higher altitudes improve productivity and area coverage. Therefore, flight altitude should be selected according to project requirements, balancing geometric quality and operational efficiency.

Advances in optical measurement technologies have increased demands for accuracy, speed, and automation in coordinate metrology. This contribution introduces CENAGIS-MET, a metrological verification standard developed at the Warsaw University of Technology for assessing active range-based systems such as terrestrial laser scanners (TLS). Unlike traditional calibration fields designed for small ranges, CENAGIS-MET enables evaluation over large measurement areas using modified VDI/VDE guidelines.

The methodology incorporates probing error, sphere-spacing error, and flatness assessment using high-precision ceramic artefacts. Tests were conducted on Leica RTC360, Leica Nova MS60, Z+F 5006h, and a handheld Livox-based scanner (Mendeye). Results show RTC360 and MS60 fully meet the 1/5 relative error criterion, confirming suitability for engineering-grade applications. Z+F 5006h achieves partial compliance, requiring careful configuration, while Mendeye exceeds permissible thresholds, limiting its use to qualitative documentation.

In the full version of the article, broader analyses will be provided regarding the accuracy of 3D shape reconstruction used for the probing error, as well as roughness and planarity assessment for evaluating the overall distribution of the reference plane

Accurate topo-bathymetric shoreline mapping and semantic segmentation of remote sensing imagery are fundamental to monitoring dynamic coastal systems, with significant implications for sustainable management, ecological preservation, and climate resilience planning. In vulnerable regions such as Lake Huron, Ontario — where sensitive ecosystems face growing anthropogenic pressures—precise delineation of land-water interfaces enables critical applications including coastal habitat mapping, sediment flux quantification, and erosion vulnerability assessment. This study presents a training-free, open-vocabulary segmentation framework that adapts frozen vision-language models (VLMs) to automatically extract shoreline features from near-infrared (NIR) imagery. By harnessing the inherent semantic reasoning abilities of VLMs, the method achieves accurate segmentation without relying on large, annotated datasets. Extensive evaluation on the Fathom Topo Bathymetric Dataset demonstrates the model's robustness across diverse nearshore environments, highlighting its applicability as a scalable solution for coastal mapping. This research underscores the potential of integrating foundational vision language models into geospatial workflows to enable automated, high-resolution environmental monitoring in data-limited settings.

The coastal zones consist of the interfaces between land and sea, undergoing the mobility of the shoreline at an unprecedented pace over the last centuries. Such a trajectory, at the global scale, exacerbates the coastal risks (intersecting hazards, exposure and vulnerability), calling upon a scalable methodology to ensure the precise and accurate monitoring. One of the observation solutions resides in the satellite platform provided with the finest spatial and spectral resolutions. Because remote sensing is a science of trade-offs, no sensors can be both excellent in spatial and spectral specificities.

We propose an original research study to create an imagery endowed with both high spatial and spectral characteristics, purposed to classify a representative coastal zone (12 habitat classes) in a temperate area in Brittany, France. The methodology highlights a transferable fusion procedure based on the simultaneous acquisition (10-min difference) of the 30-m hyperspectral PRISMA satellite imagery and the 3-m PlanetScope (Dove-R) imagery, made possible given the very high temporal resolution of the PlanetScope constellation. The spatial resolution of the hyperspectral PRISMA imagery, in the visible and near-infrared spectrum (63 bands), was successfully upscaled at 3 m, using a bandwise linear prediction from the 4 PlanetScope Dove-R bands (collected at 3 m). The model residuals showed that the pansharpened PRISMA imagery (5 m) was better enhanced (absolute deviation of 0,011) than the original PRISMA imagery (30 m, absolute deviation of 0,015). Seawater and mudflat were the best habitats upscaled, whereas the road and the roof were the worst classes predicted.

The evolution of mapping platforms has followed a consistent pattern: professional instruments are complemented by consumer devices that trade precision for scalability. Unmanned aerial systems transformed aerial photogrammetry by making it accessible beyond traditional aircraft, and smartphones equipped with RTK have demonstrated viable terrestrial mapping. This paper extends that progression to vehicle-based mapping by presenting SurveyXR, a web-based calibration and georeferencing framework that converts consumer vehicle dashcam video into photogrammetry-ready georeferenced imagery. The system addresses two technical problems: determining the geometric relationship between uncalibrated consumer cameras and a known navigation trajectory and producing per-frame exterior orientation parameters suitable for Structure-from-Motion processing. This pipeline implements checkerboard-based intrinsic calibration with automated quality diagnostics, Perspective-n-Point exterior orientation solving, and GNSS-synchronized frame extraction with lever arm correction. All computation runs in a browser or lightweight cloud backend, requiring no local software installation. The framework was tested on a 2026 Tesla Model Y equipped with PPK GNSS on the Ohio State University campus. Georeferenced frames were verified against the GNSS trajectory, confirming correct spatial positioning. The paper documents the calibration methodology, time synchronization model, and coordinate geometry, and discusses error sources and the path toward quantitative accuracy validation.

In this paper, we describe a modern, efficient, accurate and reliable stitching algorithm that JOANNEUM RESEARCH (JR) has developed for the PhaseOne PAS Pana multi-camera system. We present a new "constraint" projective transformation (CPT) approach, reducing the eight parameters of a standard projective transformation to only six, physically meanigfull parameters: Correction scale, parallax in x- and y-direction and three relative orientation angles. Based on the CPT, tie point measurements of all available image overlaps (NIR/NIR, RGB/NIR and RGB/RGB) are adjusted simultaneously within a common virtual image plane. As the CPT contains no over-parametrization any more for modelling the relative orientation of the (calibrated) camera modules we expect a more accurate and stable stitching result which will be evaluated by analysing the stitching parameters of consecutive PAS Pana shots of a flight line.

This paper presents tests of a new fully integrated multi-sensor airborne system that comprises LiDAR, multiple cameras, inertial measuring unit (IMU), GNSS, and their associated software for data acquisition, processing, integration, calibration, and map production. The technical analysis presented in this paper focuses on multi-sensor system integration that statistically addresses a multi-stream of LiDAR ranges, pixels from multiple cameras, position and orientation of each LiDAR range and each photo center derived from the GNSS/IMU trajectory. The impact of processing the trajectory in two different ways, namely: Post-Processed Kinematic (PPK using Single Base Station) and Trimble Post-Processed RTX (PP-RTX) is evaluated. Real-world data sets acquired with the Vexcel UltraCam Dragon in Austria and USA are used in this paper to address system performance in a real-world environment. Test results confirm the suitability of both approaches for trajectory processing, Single Base and PP-RTX, as well as the consistent positional accuracy of georeferencing solutions for imagery and lidar.

Matrix detectors and colour filters arrays are more widely used for satellites and rover missions in the past years. Recently, four CO3D (from “Constellation Optique 3D” in french) satellites equipped with COTS matrix Bayer sensor were launched and calibrated. Both the sensor sampling distinctive features and the new Step & Stare guidance mode are leading to new calibration and processing paradigms. In this paper, we delve into techniques dedicated for such Bayer matrix-based system, mainly but not limited to high-resolution (HR) Earth-observation (EO) satellite missions. We first describe dedicated techniques for in-orbit radiometric performance assessment like signal-to-noise ratio (SNR) and modulation transfer function (MTF). Then we address ground processing dedicated to Bayer acquisitions. Finally, we demonstrate the validity of our approach with CO3D in-orbit measurements. We also apply the radiometric ground processing on real images and provide a comparison with Pléiades-HR imagery, demonstrating the many benefits of the CO3D mission and all its novelties. CO3D in-orbit testing (IoT) is still ongoing eight months after launch, the in-flight performances are not presented in this paper due to confidentiality agreement.

This study presents a systematic evaluation of the relative geometric accuracy of UAV photogrammetry over drifting Arctic sea ice, addressing critical challenges posed by textureless surfaces and dynamic motion. Utilizing data from 18 shipborne UAV flights during the FACE2024 expedition, the research quantifies the impact of sea ice drift on orthomosaic horizontal accuracy. A methodological framework is established that incorporates shipborne GNSS data for drift correction, aligning image positions to a common reference frame under the assumption of consistent icebreaker–ice motion. Accuracy assessment is performed using onboard control and check lines of known lengths, enabling reliable relative error measurement without traditional ground control points (GCPs), which are infeasible on drifting ice.

Results demonstrate that drift velocity and total drift distance have a strong positive correlation with root mean square error (RMSE) before correction (r = 0.70 and r = 0.79, respectively), while the flight-drift angle has minimal influence (r = –0.13). The application of ship-position-based drift correction significantly improves accuracy, reducing RMSE by an average of 0.23 m and achieving a high relative accuracy of approximately 10 cm for imagery with 2–4 cm ground sampling distance. The use of control lines alone also substantially enhances results. This work validates the efficacy of drift correction and provides practical guidance for mission planning and data processing, confirming that standard UAVs and commercial photogrammetric software can produce reliable results in challenging polar environments when appropriate corrections are applied.

This paper examines how to align PlanetScope and Sentinel-2 vegetation indices, focusing on the Normalized Difference Red Edge (NDRE) index, which is commonly used in precision agriculture for prescription maps. While Sentinel-2 is popular for crop monitoring, its low spatial resolution limits use in small or irregular fields. PlanetScope provides higher-resolution, more frequent imagery, but its sensor differs from the Sentinel-2, limiting compatibility with current research and tools. By testing three adjustment methods, the study shows that it is possible to align PlanetScope NDRE values with Sentinel-2: M1 (Linear Regression + Histogram Shifting + Histogram Matching), M2 (Histogram Matching), and M3 (per-band linear regression before index calculation). Two dates from 2022 were selected as representative seasonal extremes from the broader 2021–2023 dataset of 56 image pairs (Baldin, 2025), which was further analyzed through time-series methods. Resampling direction (PS→10 m, S2→3 m) minimally affects RMSE/MAE but significantly alters spatial structure and Moran’s I values; downscaling PS to 10 m decreases Moran’s I. M2 is suitable for standard applications, whereas M3 is preferable when preservation of spatial structure is important. Across the four examined scenarios, all methods reduce RMSE below the 0.07 agronomic threshold, with calibrated RMSE ranging from 0.02 to 0.05 (up to 0.06 across the full 56-pair dataset). M3’s advantage lies in how effectively it reduces spatial autocorrelation mismatch: a 43.4% reduction in Moran’s I (versus ~18.2% with M1 and M2) in the four example scenarios, and 39.5% versus 28.4% (M1) and 28.2% (M2) reduction over the full dataset.

This study introduces an integrated airborne sensor system that combines mid-wave infrared (MWIR) imaging, a high-resolution aerial camera, and GNSS/IMU navigation for all-day, high-precision remote sensing. The MWIR subsystem adopts a Frame-scanning mechanism to achieve wide-swath and efficient thermal data acquisition. A unified calibration and synchronization framework was developed to ensure temporal and spatial consistency among sensors, including precise time synchronization, lever-arm and boresight calibration, and radiometric correction. The refined GNSS/IMU trajectory supports accurate co-registration between MWIR and optical imagery. Field experiments in China demonstrated stable system performance and consistent geometric–radiometric alignment under various illumination conditions. The integrated dataset enables detailed thermal–optical reconstruction, revealing thermal features and material contrasts not observable in visible imagery. The system supports applications such as infrastructure inspection, environmental monitoring, and emergency response. With its compact structure and modular design, the proposed platform provides a practical reference for next-generation airborne sensor integration and real-time data fusion in high-resolution mapping missions.

Existing satellite mission-planning algorithms are primarily designed for homogeneous single-payload constellations, making them insufficient for coordinating heterogeneous satellites such as optical and SAR systems. Moreover, most current approaches rely on highly abstract task models that neglect the fact that a single satellite may operate under multiple observation modes, each imposing distinct constraints on geometry, attitude maneuvering, and resource utilization. In addition, few studies have addressed the integrated scheduling of point-target and area-target missions, which is essential for scenarios combining discrete and continuous observation demands.

This study proposes an integrated scheduling algorithm for multi-mode, multi-scenario, and multi-task Earth-observation constellations. The algorithm formulates mission planning as a unified spatiotemporal optimization problem, jointly considering visibility, sensor compatibility, attitude feasibility, and onboard resource limits. A CDCL-enhanced constraint-programming solver is employed to enable coordinated scheduling across different observation modes and target types.

Experimental validation on hydropower and disaster-monitoring scenarios shows that the proposed method significantly improves coverage, cross-sensor synergy, and responsiveness compared with traditional homogeneous schedulers. The results establish a new paradigm for integrated and intelligent mission planning of heterogeneous, multi-mode satellite constellations.

Navigating Unmanned Aerial Vehicles (UAVs) in Global Navigation Satellite System (GNSS)-denied environments requires reliable autonomous localization techniques. This study proposes a vision-based localization framework utilizing satellite true orthophotos and Digital Surface Models (DSMs) as absolute geospatial references. The algorithmic pipeline integrates deep learning architectures—specifically SuperPoint and LightGlue—to establish robust image-to-map feature correspondences. The matched correspondences are used to estimate camera exterior orientation parameters through collinearity-based spatial resection with an Iteratively Reweighted Least Squares (IRLS) approach. To validate the proposed methodology, a multi-altitude dataset (100–250 m) was acquired across structurally diverse terrains, including dense building, high vegetation, and bare ground areas. Experimental evaluations demonstrate that the framework achieves meter-level absolute positioning accuracy and stable pose estimation. Analyses further reveal that matching robustness and localization success rates depend heavily on terrain texture and flight altitude; geometrically structured urban scenes at moderate-to-high altitudes consistently yield reliable correspondences, whereas low-texture environments and lower flight altitudes present persistent challenges for continuous visual tracking.

Recent handheld scanners increasingly integrate geometric (LiDAR-based) and visual (image-based) SLAM (Simultaneous Localization And Mapping), promising low-cost and flexible solutions for surveying tasks. This paper evaluates the accuracy of three such systems: the XGRIDS Lixel K1, the SHARE S20, and a Pix4D solution pairing an iPhone Pro with an Emlid Reach RX GNSS (Global Navigation Satellite System) antenna. We conducted experiments in two distinct environments: Scene 1, with continuous, high-quality RTK (Real-Time Kinematic) coverage, and Scene 2, which included an indoor trajectory resulting in a temporary loss of the RTK fix. Accuracy was validated against independent GNSS check points. In Scene 1, the Pix4D solution delivered survey-grade results, achieving a RMSE (Root Mean Square Error) below $3~\text{cm}$ in the $X, Y$, and $Z$ directions. The XGRIDS and SHARE scanners yielded larger maximum errors, around $10~\text{to }15~\text{cm}$. In Scene 2, accuracy degraded; the Pix4D solution's maximum error increased to approximately $12~\text{cm}$ , while the Share S20's maximum error exceeded $25~\text{cm}$. We conclude that while the fusion of visual and geometric SLAM is powerful, a stable RTK fix remains critical for achieving consistent survey-grade accuracy with current low-cost handheld scanners

Stem diameter is a key parameter for assessing woody vegetation growth and its ecological and economic benefits, including biomass production, carbon sequestration, and urban ecosystem services. Recent advances in handheld mobile laser scanning (HMLS) enable efficient acquisition of high-density point clouds for deriving tree structural attributes in complex environments. This study presents an automated workflow for tree detection and diameter at breast height (DBH) estimation in an urban park, using two HMLS systems: the GoSLAM RS100i and the FJD Trion S1. The influence of point cloud density and subsampling resolution (2 - 4 cm) on detection and accuracy was evaluated. Reference data for 69 trees were collected using a forestry caliper and total station, while HMLS datasets were georeferenced with RTK-GNSS. The workflow included point cloud filtering, terrain modelling, stem extraction, and DBH estimation through cylindrical fitting. Detection performance differed between systems and was strongly affected by point density. The GoSLAM RS100i detection rate decreased from 97.1% at 2 cm to 53.6% at 4 cm spacing, whereas the FJD Trion S1 maintained stable performance (~87%) across all resolutions, likely due to higher point density. DBH estimation accuracy was similar for both systems, with RMSE values of 3.3–3.6 cm for filtered data and up to 4.9 cm when all detections were included, alongside a consistent positive bias (1.7–2.5 cm). Subsampling had no significant effect on DBH accuracy, indicating robustness to moderate density reductions. Overall, HMLS systems provide reliable DBH estimates in urban environments, with performance mainly influenced by point cloud quality.

In general, the obstacle detection systems mainly rely on depth cameras or AI-based vision approaches; however, these methods are often constrained by limited fields of view and the need for continuous model retraining to adapt to complex and dynamic industrial scenes. To overcome these limitations, this study proposes a LiDAR-based obstacle detection and field monitoring system integrated with a quadruped robot. The proposed system focuses on three main components: real-time field mapping, intelligent path planning with obstacle avoidance, and field change detection. LiDAR point cloud data are pre-processed using pass-through and voxel grid filters, followed by coordinate transformation into the robot reference frame. The Cartographer simultaneous localization and mapping (SLAM) algorithm are employed to generate high-resolution occupancy grid maps for navigation. For autonomous operation, erosion processing and connected component labelling are used to define safe regions, while the A* algorithm enables efficient path planning and adaptive obstacle avoidance in complex environments. To detect unknown obstacles and environmental changes, Gaussian filtering and map differencing are applied, and map similarity is evaluated using histogram analysis and SIFT-based feature matching. Experimental results demonstrated that the system achieves a mapping resolution of 0.05 m and satisfies the Taiwan Association of Information and Communication Standards (TAICS) requirements, including 0.2 m planimetric accuracy and less than 0.1 m positional error. The proposed approach effectively identifies unknown obstacles and visually highlights risk areas, providing a reliable solution for intelligent workplace safety monitoring.

Centroid extraction from star images is a critical component in achieving high-precision satellite attitude determination. Prevailing approaches primarily focus on suppressing a single type of error or depend on fixed filtering and compensation parameters, often lacking a multidimensional and fine-grained analysis and handling of diverse error sources. To address these limitations, this paper proposes a compensation method for centroid extraction based on error classification and modeling, coupled with an adaptive strategy selection mechanism to improve accuracy. Experimental results demonstrate the efficacy of the proposed method: on a set of 30 to 300 laboratory-simulated star images, it enhanced the average centroid extraction accuracy from a baseline of 0.31–0.45 pixels to 0.11–0.19 pixels when using a Static model Unscented Kalman Filter (UKF) integrated with four sub-pixel interpolation techniques. Furthermore, for a larger dataset of 300 to 600 star images simulated at a 300 Hz frame rate, the method achieved an accuracy improvement exceeding 50% across five different motion model UKF methods, demonstrating robust performance.

In recent years, Space-Air-Ground sensing data has become increasingly abundant. This paper focuses on the technology system of integrated intelligent sensing in Space-Air-Ground remote sensing, aiming to integrate data from different platforms and sensor types through deep collaboration to meet the growing demand for high-precision, high-frequency, and near-real-time monitoring in scenarios such as land change detection, natural resource development, and disaster emergency response. This paper constructs a technical framework for the integrated intelligent sensing technology system for Space-Air-Ground remote sensing,focusing on core technical methods such as multi-source data governance and correlation, component-based AI interpretation model development, and the construction of application agents based on multi-modal large models.

This study validated the application of an integrated space-air-ground intelligent monitoring system through a typical ecological restoration project monitoring and supervision case. A test area of approximately 5.61 hectares in Hunan Province, China, was selected to construct an ecological restoration monitoring agent. This agent comprehensively utilized multi-temporal satellite imagery spanning eight years, UAV image data, and tower-based videos.Driven by natural language instructions, the agent autonomously planned task chains, coordinated multi-source data, and triggered models. After the implementation of the ecological restoration project, the results showed 4.87 hectares of new grassland and 0.74 hectares of new forest within the area, achieving intelligent identification and quantitative, automated assessment of land cover types, restoration progress, and ecological recovery outcomes. The experiment demonstrated the system’s advantage in "rapid identification and early warning",forming an intelligent operational closed loop of "monitoring-analysis-decision-feedback."

Traditional pavement inspection and data collection are often constrained by traffic conditions, operational safety, and equipment costs, making it difficult to achieve both efficiency and large-scale coverage. To address these limitations, this study employs a Pavement Roughness Index and Distress Extraction System (PRIDEs), which integrates high-resolution industrial cameras, high-precision global navigation satellite system (GNSS), wheel pulse sensors, and an onboard computer to acquire high-quality images under high-speed driving conditions. Using photogrammetry and computer vision techniques, camera poses are reconstructed to generate dense point clouds, digital surface models (DSMs), and orthophotos for detailed pavement distress analysis. However, the acquired imagery is affected by dynamic shadows and lens-focusing induced blur, resulting in ghosting artifacts and inconsistent orthophoto quality. To mitigate these issues, this study proposes a masking strategy during orthophoto generation, where U-Net is employed to detect shadow regions and Laplacian variance is used to identify blurred areas. By integrating these masks, more uniform and higher-quality orthophotos can be produced. Experimental results demonstrate that the proposed approach effectively reduces false positives and false negatives of crack detection caused by shadows and blur, thereby improving the reliability of orthophotos for automated pavement condition assessment.

Ultra-Wideband (UWB) technology is gaining

attention for indoor positioning due to its high accuracy, low

latency and resilience to interference, making it ideal for

environments where GNSS (Global Navigation Satellite System)

signals are unavailable—such as warehouses, hospitals, and

underground facilities. However, UWB systems can suffer from

reduced accuracy under Non-Line-of-Sight (NLOS) conditions

and dynamic deployments. This paper proposes a novel bias

aware EKF (Extend Kalman Filter) model, combined with

Anchor Self-Localization method for localization in indoor

environments, and enhancing the flexible deployment of

anchors. The proposed model demonstrates an overall

improvement of 32% and 41% in positioning accuracy

compared to traditional methods across both indoor and

outdoor environments respectively. The paper demonstrates the

proposed ASL method, it performs at par with conventional pre

calibrated methods where anchors are to be localized manually.

Together, the Bias-Aware filtering and ASL approach enhance

the scalability and reliability of UWB-based Indoor Positioning

Systems (IPS) for real-world applications.

Handheld 3D scanners have gained increasing importance in recent years due to their flexibility and declining acquisition costs. While high-end systems provide standardized accuracy specifications, affordable devices often lack reliable and comparable benchmarks. This paper evaluates the geometric accuracy of three low-cost handheld 3D scanners (Revopoint Pop 3 Plus, Revopoint MetroX, 3DMakerPro Moose 3D scanner) compared to two high-end systems (Hexagon MARVELSCAN, Hexagon Absolute Arm with AS1 scanner), using the ZEISS Atos 5 structured light system as reference. Five different test objects with varying material and geometric properties were used for practical assessment. Results reveal significant differences regarding flatness, detail fidelity, and robustness: while some low-cost scanners achieve remarkable accuracies, their performance is less stable under varying conditions. High-end systems, in contrast, consistently provide high precision and reproducibility. This study provides a well-founded classification of current handheld 3D scanners and practical guidance for their application in science, industry, and education.

In recent years, LiDAR-based localization has been widely explored. Among them, Scene Coordinate Regression (SCR)-based

methods have demonstrated outstanding accuracy and robustness in city scenes. Integrating these models with traditional Simultan

eous Localization and Mapping (SLAM) methods is expected to enhance localization accuracy and reliability further. This paper

proposes loosely coupled fusion methods integrating an SCR model with SLAM to improve localization accuracy and robustness.

The approach addresses the information loss problem in high-level sensor fusion while maintaining computational efficiency. The

method achieves tighter data association and complementary performance advantages by strategically combining LiDAR-based

localization results with SLAM pose estimates. Experimental results in the NCLT and HeLiPR datasets demonstrate that the pro

posed fusion framework effectively corrects SLAM drift and maintains stable pose estimation accuracy under diverse environmental

conditions. Furthermore, the sparse-frame coupling strategy significantly reduces computational overhead without degrading local

ization performance, making the method suitable for practical applications. The system exhibits improved robustness across regions

and LiDAR configurations while preserving real-time operation capabilities.

Matrix detector are becoming increasingly common in optical imaging satellite. To maintain good geometric quality of the images, the line of sight (LOS) of each pixel must be known precisely. This paper aims to estimate the performance of our method of LOS calibration on Co3D datas, which requires a precise geometric model for altimetric reconstruction.

In cold-climate regions like Canada, pavement infrastructure deteriorates rapidly due to extreme freeze-thaw cycles and heavy use of de-icing salts, accelerating the formation of structural cracks and imposing a financial burden on municipal budgets. By providing an automated LiDAR (Light Detection and Ranging)-based detection framework, this research offers a cost-effective, high-precision monitoring tool that enables early intervention, reducing long-term repair costs and enhancing road safety across Canadian provincial networks. This study evaluates the performance of Support Vector Machines (SVMs) and Multi-Layer Perceptrons (MLPs) for crack classification in a multi-dimensional feature space. We propose integrating geometric height (H) with a novel set of radiometric indices, including the Normalized Difference Intensity Index (NDII) and the Green Ratio (GR), to enhance classification stability. Results demonstrate that both SVM and MLP achieved comparable accuracies of 87% and 86%, respectively, in low-dimensional feature spaces. A critical analysis of the MLP learning curves reveals that the introduction of NDII acted as a numerical stabilizer, mitigating the oscillations caused by raw brightness fluctuations. Furthermore, the study identifies an information ceiling, as architectural expansion of the MLP improved convergence stability but did not exceed the 87% accuracy threshold. These findings provide a robust framework for automated road maintenance using stabilized radiometric features in LiDAR-based distress identification.

The aim of the study was to propose a procedure enabling accurate mapping of the above water and underwater areas of the bridge. The object of the study was a road bridge located approximately 30 km north of Warsaw, Poland. The bridge is 332 m long and 13.5 m wide. The bridge is located over Lake Zegrze. A mobile topographic Norbit iLIDAR system was used to measure the bridge structure above the water surface. Bridge pillars measurement and the shape of the bottom of the water reservoir in the immediate vicinity of the bridge were performed using a Norbit Winghead i77h multibeam echo sounder. In order to bridge point clouds integration, a workflow has been proposed: LIDAR and MBES data filtering, consideration of the speed of sound in water, LIDAR and MBES data calibration including Patch Test and MBES cross check. As a result, the integrated point cloud of the bridge was created. The LIDAR point cloud resolution was 1 cm and the MBES point cloud resolution was 0.02 m. The created point cloud of bridge provides could be useful for monitoring erosion and accumulation phenomena, analyzing the stability of bridge pillars and verifying hydrodynamic models.

Accurate characterization of the terrestrial laser scanner (TLS) beam footprint is essential for understanding the stochastic behavior of the scanner. Estimating the laser footprint width is crucial for determining the correlation length between neighboring observations, and thus for providing a realistic estimation of the correlation elements within the variance-covariance matrix of the measurement uncertainties. In this contribution, an intensity-based workflow is introduced to estimate the laser beam footprint by deriving the line spread function (LSF) from the edge spread function (ESF). The proposed method applies no assumptions regarding the geometric or physical behavior of the beam, allowing the footprint to be determined directly from the measured intensity values. Using the BOTA8 target, the Z+F Imager 5016A was investigated at two different distances and two scanning rates with varying mirror rotation speeds. The results provide insights into the influence of distance and scanning rate on the laser beam footprint.

UAV-borne PV point clouds often suffer from severe shadow artifacts and color dropouts, limiting their use for reliable inspection and digital twin construction. We introduce a fully vectorized color rectification framework that exploits panel symmetry and 1D signal processing to restore a consistent radiometric appearance. Starting from a segmented solar-panel point cloud, the method first normalizes panel geometry via RANSAC-based plane segmentation and rotation to a canonical xy-plane, then extracts base colors by clustering RGB values to identify “panel blue” and “grid white” regions. It subsequently detects grid parameters by projecting filtered grid and panel points into 1D spatial density histograms along the x- and y-axes to estimate spacing, offset, and grid-line thickness, and finally performs vectorized recoloring and color remapping of grid and panel points using the recovered parameters. By decoupling periodic grid structure from illumination noise, our approach achieves visually near-perfect color restoration while eliminating intra-semantic variance across modules. The resulting high-fidelity, shadow-free point clouds provide a mathematically consistent foundation for PV digital twins and automated asset evaluation.

EarthDaily Constellation (EDC) is a ten-satellite, sun-synchronous mission optimized for persistent, daily monitoring of global land and designated coastal waters. Each spacecraft carries co-aligned VNIR, SWIR, and TIR imagers and acquires nadir-only imagery at ~10:30 LTAN to stabilize collection for optimal change detection. A systematic acquisition plan builds a global spatiotemporal archive; EarthPipeline performs automated geolocation, orthorectification, atmospheric correction, QA, and wide-area compositing. Bands and metadata are designed for CEOS CARD4L-SR alignment and inter-sensor interoperability with Landsat and Sentinel-2. The talk reports early on-orbit performance—geometric accuracy, radiometric stability—and benchmarking of atmospheric correction and cloud/shadow masking against ESA Sentinel-2 processing, with a focus on time-series consistency for analytics and ML. We also outline specialized applications concepts and readiness.

This paper presents a static calibration approach for lightweight laser scanning systems, utilising planes as control entities, with a focus on estimating boresight misalignment angles. The calibration with the system static, aims to minimise errors originating from several sources, such as position and attitude systems, time synchronisation, and control features measurement. The mathematical model is based on the plane equation, combined with the equations of laser scanning. The estimation is performed with the combined model of least squares. Experiments in a terrestrial calibration field were performed. The results show that the approach successfully estimates the boresight misalignment angles, reducing the errors of the point cloud with respect to the control planes.

Monitoring water resources is essential for supporting human activities and enabling informed decision-making. Since its launch in 2022, the Surface Water and Ocean Topography satellite mission has provided global observations of surface water elevation for rivers, lakes and oceans. Several studies have evaluated SWOT performance for ocean applications (Hay et al., 2025, Lichtman et al., 2025) and continental water bodies (Patidar and Indu, 2025). However, no comprehensive assessment has yet focused on Canadian inland waters. This research presents an initial evaluation of SWOT water surface elevation observations using hydrometric stations operated by Water Survey of Canada (WSC). This evaluation covers the period between operational orbit reached in July 2023 and December 2025.

This presentation introduces a novel approach for high-resolution Earth observation using GNSS-SAR imaging with spaceborne GNSS-Reflectometry satellites. By leveraging low-level intermediate frequency (IF) signals from the CYGNSS satellite constellation, our work demonstrates the feasibility of forming GNSS-SAR images from spaceborne GNSS-R data. The integration of advanced weak signal tracking algorithms and tailored SAR image formation techniques enables the retrieval of Earth observation data with unprecedented spatial and temporal resolution. This addresses longstanding challenges in space-based GNSS-R remote sensing, such as limited spatial resolution and weak signal reception. The LEO satellite-based GNSS-SAR approach offers significant advantages, including global coverage, rapid revisit times, and the potential for onboard processing. These features collectively support scalable, near real-time monitoring of dynamic Earth processes, making this technique highly relevant for extreme weather surveillance, disaster preparedness, and environmental monitoring.

This study presents a low-cost LiDAR–IMU–GNSS mapping framework for continuous and globally consistent three-dimensional reconstruction across indoor–outdoor environments. The work addresses a key limitation in current SLAM and GNSS-integrated systems, where LiDAR-based approaches provide strong local geometric accuracy but lack reliable global referencing, while GNSS-based solutions often rely on high-precision corrections such as RTK or PPP, limiting scalability and deployment in urban environments.

Building upon the Dense Multi-Scan Adjustment SLAM (DMSA-SLAM) framework, the proposed system introduces a structured integration of standalone Single Point Positioning (SPP) GNSS through an external alignment strategy, ensuring that global referencing is achieved without compromising locally consistent LiDAR–Inertial geometry. The framework further incorporates explicit multi-level structural constraints to support consistent cross-floor reconstruction, along with a bounded optimization and loop closure strategy that maintains stability and prevents global trajectory deformation without requiring full pose graph optimization.

The system is validated in a multi-storey urban building under challenging GNSS conditions, including complete signal outages and urban canyon effects. Results demonstrate sub-decimeter indoor geometric accuracy and meter-level global georeferencing using low-cost sensors. Comparison with a high-accuracy terrestrial laser scanning (TLS) reference confirms reliable reconstruction quality, while the proposed system achieves rapid mapping in a single continuous trajectory using a significantly lower-cost sensor suite.

Overall, the framework provides a practical and scalable solution for infrastructure-free indoor–outdoor mapping, supporting applications in BIM, digital twins, and urban asset management.

This contribution presents a comprehensive calibration framework for a hybrid solid-state LiDAR system designed for small body exploration. Integrating imaging modeling, photon-count–based parameter estimation, and multi-scale ground experiments, the method effectively corrects pixel-dependent range and angular errors. Rigorous validation demonstrates centimeter-level accuracy in both mapping and navigation modes, confirming the framework's robustness and its critical role in enhancing deep-space mission capabilities.

This research introduces a deep learning-based, multi-sensor framework for automated damage detection in cultural heritage structures using fused thermal and optical imagery. Conducted across five historic sites in South Delhi, India, the study targeted common degradation forms—cracks, spalling, and biological growth—through high-resolution image acquisition using a FLIR T1030sc thermal camera and RGB sensors. Fused datasets (MXS and thermal-optical blends) significantly outperformed optical-only inputs, with the YOLOv11-Tuned model achieving a peak mAP of 91.8%. The fusion allowed reliable detection of subsurface anomalies and fine-scale damage often missed by traditional visual inspections. Oriented Bounding Box (OBB) variants improved localization of non-linear features, while genetic algorithm-based hyperparameter tuning enhanced model precision. The framework offers a scalable, non-invasive, and accurate alternative to manual inspection, supporting early diagnostics and long-term conservation planning. This approach demonstrates the transformative potential of AI and remote sensing in preserving architectural heritage against both environmental and anthropogenic threats.

Hyperspectral data alone in mineral exploration often suffers from limitations including signal noise, coarse spatial resolution, and spectral variability, which can hinder mineral discrimination. To address these challenges, we integrate Short-Wave Infrared (SWIR) and Visible Near-Infrared (VNIR) hyperspectral data cubes with complementary sensor modalities, including RGB imagery and LiDAR acquired from indoor scans of drilled core. This multi-sensor fusion enhances the reliability and accuracy of mineral maps by leveraging the strengths of each modality.

At GeologicAI, our indoor scanning platform captures multi-modal data from a box of core using a variety of different sensors. A critical preprocessing step involves isolating the drilled core from the background. We further applied a continuous wavelet transform (CWT) for a scalogram analysis enables the differentiation of unclassified spectra based on their frequency-scale characteristics. Following spatial masking and unclassified spectral filtering, we apply a local end-member selection regime utilizing RGB, VNIR and SWIR for all valid pixels. Afterwards, non-negative least squares (NNLS) linear unmixing. While SWIR remains the primary source for mineral identification and abundance calculations, VNIR and RGB data provide critical support in resolving ambiguities either confirming the presence of minerals difficult to detect with SWIR alone or excluding candidates based on VNIR disagreement or RGB colour disagreement. Mineral maps derived from SWIR data exhibit a reconstruction residual error of 12.4%. While the integration of VNIR data does not necessarily reduce this residual, it enhances confidence in abundance estimations, particularly in regions where SWIR alone cannot separate end members.

.

Calibration of terrestrial laser scanners (TLS) is paramount for ensuring high-precision measurements. The costs and efficiency of calibration pose significant challenges for both instrument manufacturers and end-users conducting self-calibration of TLS systems. To date, there has been a lack of theoretical methods for quantitatively analyzing and optimizing the geometric network of targets within calibration fields. This study proposes the TNet-GDOP (Target Networks Geometric Dilution of Precision) theory and its mathematical model to quantitatively evaluate the impact of target distribution on parameter solution precision. We propose the optimized the target network strategy based on the precision contribution factor of TNet-GDOP (OptimizeTNet-PCF), a target distribution optimization algorithm with a well-defined scoring function. OptimizeTNet-PCF can reduce the number of targets with minimal effect on parameter precision while suppressing anomalous observations. The number of targets was reduced to one-eighth (from 140 to 16), with ranging parameter variations less than 0.1 mm and angular parameter variations less than 0.2″.

This paper examines the feasibility of calibrating a consumer camera with a calibration panel to accurately reconstruct seabeds in shallow water. Specifically, it assesses whether calibration parameters determined based on the panel can be applied to an independent set of images captured under different conditions. The study also examined the effect of the analyzed approach on the final accuracy of the point cloud. The analysis covered three calibration variants: (1) external calibration based on an underwater panel, (2) preliminary calibration in which the panel parameters were used as initial values for further optimization, and (3) fully automatic autocalibration. The results showed that calibration using the panel does not improve reconstruction quality and can lead to model distortion. The highest accuracy was achieved with in situ autocalibration, supported by underwater control points.

The Advanced Land Observing Satellite-4 (ALOS-4), launched on July 1, 2024, observes the Earth's surface using its onboard Phased Array type L-band Synthetic Aperture Radar (PALSAR-3). Japan has continuously advanced L-band radar technology, and ALOS-4 offers significantly improved observation performance compared to its predecessor, PALSAR-2, aboard ALOS-2, which was launched on May 24, 2014. ALOS-4 is designed to achieve both high spatial resolution and a wider observation swath—expanding the 3 m strip map mode coverage from ALOS-2’s 50 km to 200 km.

By employing this wide-swath observation capability, ALOS-4 can acquire 3 m dual-polarization data over Japan approximately once every two weeks. These frequent observations support disaster management by providing timely information on events such as volcanic activity, land subsidence, and landslides. Moreover, the high-temporal-resolution 3 m dual-polarization data are valuable for a wide range of applications, including agriculture, ocean monitoring, and environmental studies.

To effectively utilize ALOS-4 data, it is essential to integrate it with the long-term archive of ALOS-2 observations, enabling time-series change detection. Maintaining consistent geometric and radiometric quality between ALOS-2 and ALOS-4 data through cross-calibration and validation is therefore critical. This paper presents the results of these efforts and outlines the current use of ALOS-2 and ALOS-4 data under the ALOS-2 Public–Private Partnership (PPP) Phase B activities.

This study investigates how differences in UAV-LiDAR sensor performance affect the accuracy of canopy radiative transfer simulations. Conducted in a Japanese larch forest in Yamanashi, Japan, the research compares two UAV-mounted LiDAR systems—YellowScan Explorer and Voyager—flown over the same plot. The simulation approach uses a voxel-based model to estimate solar irradiance attenuation and reflection, optimizing parameters to match Sentinel-2 NIR reflectance.

Results show that Voyager, which produced over twice the point density of Explorer, achieved a higher correlation with Sentinel-2 data (r = 0.74 vs. r = 0.67). This suggests that higher point density improves upper-canopy representation and enhances simulation accuracy. However, the study also emphasizes the continued importance of complementary ground-based LiDAR (e.g., handheld or TLS) for capturing understory structure.

The findings highlight that UAV-LiDAR is essential for accurate canopy modeling, but sensor specifications—particularly point density—significantly influence radiative transfer outcomes. Future work should explore integrating multiple LiDAR sources and testing scalability across diverse forest types and phenological stages.

Soil erosion is a significant threat to ecosystem quality, and the development of accurate models to map erosion susceptibility is essential for enhancing public mitigation policies. This study investigates the applicability of the algorithms Weighted Subspace Random Forest (WSRF), Random Rotation Forest (RRF), and Naive Bayes (NB) to map soil erosion susceptibility in the Rio Pardo watershed, located between the states of São Paulo and Minas Gerais. A total of 120 sample points of erosion and non-erosion sites were used, identified through high-resolution images from Google Earth Pro and field visits. Fifteen conditioning factors were initially considered, but after analyzing multicollinearity and factor relevance, only thirteen were selected for the final modeling. The dataset was randomly divided into 70% for training and 30% for testing to assess the robustness of the models. The performance of the algorithms was evaluated using metrics such as accuracy and AUC-ROC. The accuracies obtained were 0.87 for NB, 0.89 for RRF, and 0.88 for WSRF, while the AUC-ROC values were 0.93, 0.96, and 0.95, respectively. RRF showed the best performance, confirming the usefulness of these models in sustainable management and conservation of areas susceptible to erosion.

This work combines in situ measurements of near-surface air temperature with the CMCC VHR-REA_IT climate reanalysis dataset to assess the spatial and temporal dynamics of heat wave (HW) events and evaluate heat stress (HS) conditions across Italy for the period 1981-2024. HWs are characterised in terms of their frequency, duration, and intensity, while HS is evaluated through ad-hoc indices, including Humidex. A trend analysis is performed to investigate the temporal trends of HWs and of hazardous HS conditions. Results indicate a significant increase in the number of HW events alongside a growing frequency of severe thermal discomfort conditions (up to 6 days more per decade). Overall, this work underscores the intensification of heat-related hazards in the study area, emphasising the need for mitigation and adaptation strategies. The ultimate goal is to develop a scalable, open-source methodology that enables continental-scale assessments of heat extremes and their impacts.

Current Historical Geographic Information (HGI) research faces significant challenges in integrating multi-source heterogeneous data (China Historical GIS Project, 2025). The lack of unified semantic standards, effective interoperability mechanisms, and systematic organization of historical sources has led to severe "data silos." Consequently, a core problem remains: the semantic fragmentation, temporal inconsistency, and disconnected evidence chains of complex cultural landscape data (Southall, 2014). While existing approaches successfully utilize traditional GIS for spatial management or foundational ontologies (e.g., CIDOC CRM) (Bekiari et al., 2024) for static artifact cataloging, they struggle to formalize and compute the dynamic evolution of heritage sites over long historical trajectories.

To overcome these bottlenecks and advance the multidimensional application of cultural landscape heritage data, this study proposes a data organization framework centered on semantic normalization and standardization. Driven by a novel hybrid semantic architecture, we construct an extensible semantic foundation and a multi-source fusion mechanism. This approach seamlessly couples macroscopic cultural landscape heritage event-centric modeling with microscopic temporal annotations, strictly regulated by a "policy–ontology–rules" constraint mechanism.The framework is designed to support computable, searchable, and inferable unified knowledge representations, thereby enabling deep integration of spatio-historical big data, semantic reasoning, and evidence- based decision-making for cultural landscape heritage management.

This presentation addresses our profession’s need for new methods to identify rapidly the prominent patterns concerning the locations, time intervals, classes, and transitions that account for gross changes during sequential time intervals in a series of maps, as opposed to popular methods that compute merely the sizes of classes at time points. Trajectory Analysis is a method that computes various components of change during a time series for exactly one land cover class. Our method of Change Components Analysis extends the concepts of Trajectory Analysis to present new concepts to address multiple classes using our new free software. Our novel methods are especially effective at identifying where, when, which classes, and which transitions demonstrate suspicious changes that warrant attention to data quality. Our new methods identify also change components that can give insights to landscape processes.

Spatial association reveals the interconnected nature of geographical phenomena, describing the interactions and influences of environmental variables across geographic space. Path analysis can explore complex causal relationships between variables by analyzing path coefficients. However, in large-scale studies, path analysis methods are often affected by local effects, which can influence the accuracy and reliability of the results. This study develops a local pathway association (LPA) model to analyze local effects of pathways among variables that integrate path analysis and local pathway coefficient estimations. The LPA model was employed to investigate the spatial heterogeneity of spatial associations between factors such as climate, soil, and vegetation on the Tibetan Plateau. Results indicate that the LPA model effectively reveals the spatial variation characteristics of local path coefficients between geographic variables, avoiding the underestimation or overestimation of global path coefficients in traditional path coefficient studies. The developed LPA model provides an effective technical tool for revealing spatial differences in path associations of large-scale spatial studies. The strong data compatibility of the LPA model allows for broad applicability across various disciplines and a deeper understanding of localized interactions and variations in complex geospatial and Earth systems.

This Contribution presents a generative Flow Matching Framework for sub-daily Wildfire Progression Monitoring from combined MODIS, VIIRS and Sentinel-3 Observations. The Approach treats all available Multi-Sensor Looks as irregular Samples along an underlying Spatio-temporal Fire Trajectory and learns continuous Vector Fields that map coarse Reflectance Observations to Sentinel-2-like Reflectance and Burned Area Masks. The Input Constellation uses MODIS Bands 1, 2 and 7, VIIRS I1-I5 and Sentinel-3 OLCI Bands Oa08 and Oa17 together with SLSTR Band S6, providing complementary Information in the visible, NIR, SWIR and Thermal Domains as well as staggered Overpass Times. Labels are derived from Sentinel-2 Surface Reflectance and Burned Area Polygons from the National Burned Area Composite as well as additional manually interpreted Fire Perimeters. We expect the learned Model to reconstruct Fire Progression at 3-6 Hour Resolution for many large Events, to improve Burned Area Delineation over single Sensor Baselines, and to provide Ensemble-based Uncertainty Estimates that highlight ambiguous Regions under Smoke or Cloud. The resulting Multi-Sensor Dataset and trained Model are intended as reusable Resources for future Research on Wildfire Monitoring and Data Assimilation.

This text outlines an approach to achieving practical geospatial data interoperability through incremental, data-driven standardization rather than relying on a single, universal standard. It frames interoperability as an evolving process in which data models, syntactic formats, semantic vocabularies, and protocol bindings are progressively aligned, generating network effects that lower implementation costs. The AMPLIFY-EDS project applies these principles to the end-to-end lifecycle of Near Real Time (NRT) environmental sensor data across the UK Environmental Data Service (EDS). Led by the British Oceanographic Data Centre (BODC), the project establishes a federated API ecosystem using the OGC SensorThings API (STA), integrating multiple MQTT data streams from research vessels and partner data centres. A Python relay application performs ingestion, validation, and quality control before posting data to a FROST server, while a React frontend provides visualisation. Metadata harmonisation required community agreement on minimal entity requirements, vocabularies, and JSON schemas, drawing on schema.org and SOSA. The team then enriched STA outputs by mapping JSON to JSON-LD and creating context files validated through OGC Building Blocks.

This study presents a spatiotemporal modeling framework for predicting hourly NO2 concentrations in Seoul by incorporating dynamic vehicle activity data recorded from Digital Tachographs (DTG). Conventional Land Use Regression (LUR) models rely on static spatial predictors and therefore struggle to represent short-term emission dynamics driven by rapidly changing traffic conditions. To overcome this limitation, this research integrates high-frequency DTG variables—vehicle speed, acceleration, braking events, and truck activity—into a dynamic LUR model and evaluates hourly NO2 variability across the urban environment.

Model performance was assessed using panel regression with random effects and hourly time indicators to capture temporal fluctuations at fixed monitoring locations. The DTG-integrated model exhibited substantially improved explanatory power, raising the within R2 from 0.17 in the static baseline to 0.25. The consistent significance of DTG-derived predictors highlights the dominant influence of real-time traffic behavior on short-term pollution levels and confirms the value of incorporating high-resolution mobility data.

Hourly prediction maps revealed strong diurnal patterns, with concentrations lowest at 4 a.m. and highest at 8 p.m., when evening congestion produced values nearly double those of early morning. A LISA cluster analysis further showed that high–high spatial clusters expanded from 17% to 28% of the study area during peak hours, demonstrating increased spatial concentration of pollution. The transition of grid cells between cluster categories also indicated dynamic shifts in spatial patterns throughout the day.

Overall, this study demonstrates that integrating DTG data substantially improves the characterization of hourly pollution dynamics and provides a foundation for time-sensitive, location-specific air-quality management strategies.

This research presents a new Multi-LoD4 Application Domain Extension (ADE) for CityGML to improve the integration of Building Information Modelling (BIM) and Geospatial Information Systems (GIS) in Urban Digital Twins (UDTs). The proposed approach extends CityGML’s Level of Detail concept to better represent both exterior and interior elements of buildings while keeping their semantic information complete. It links the geometric model to a graph-based database that stores and connects all building components, allowing users to visualize and query the data interactively in a web environment. The Multi-LoD4 ADE enhances interoperability, semantic richness, and data accessibility, providing a more comprehensive and practical foundation for future UDT applications in areas such as building management and urban analysis.

This work presents an automated and adaptive pipeline for detecting photovoltaic (PV) systems in high-resolution satellite imagery. The solution was developed to support large-scale energy monitoring efforts in the state of Minas Gerais, Brazil, where geographic diversity and visual variability pose significant challenges to accurate PV identification. The proposed pipeline operates from a single pair of geographic coordinates, automatically defining the area of interest, acquiring a basemap image, classifying the spatial context through HSV histograms, UMAP dimensionality reduction, and K-Means clustering, and dynamically selecting the most suitable deep learning segmentation model. Multiple U-Net architectures with different ImageNet-pretrained encoders were evaluated to segment PV panels, and building footprints from public datasets were used to refine detections through geospatial segmentation (SamGeo). Experimental results indicate that model performance varies across environmental contexts, highlighting the importance of context-aware model specialization. Preliminary evaluations show that dynamically assigning models such as ResNet50 and VGG16 to their optimal clusters improves segmentation accuracy. Overall, the proposed methodology demonstrates a modular, scalable, and context-adaptive approach for PV system detection, suitable for integration into urban and energy monitoring platforms.

Clustering methods are widely used in regionalisation research to identify spatial and functional structures within complex economic systems. Yet different clustering specifications can lead to contrasting interpretations of regional patterns. Advanced Producer Services (APS), i.e., specialised, knowledge-intensive business services, provide a useful setting to examine these methodological choices.

This paper develops a framework comparing spatially constrained and unconstrained clustering for delineating APS employment regions in the UK. Spatial methods group neighbouring units to preserve geographic contiguity, while non-spatial methods group areas with similar employment profiles regardless of location. We ask to what extent APS regionalisation follows spatial contiguity versus functional--economic linkages that transcend geography.

Our contribution is twofold. Substantively, we show that APS in the UK form functionally coherent but spatially fragmented regions, challenging planning approaches that assume contiguous blocks of territory. Methodologically, we quantify the trade-off between cluster quality and spatial interpretability, providing a simple diagnostic to guide method choice in regionalisation studies.

Effective disaster response requires rapid allocation of limited human and material resources to dispersed and dynamically changing demands. This study formulates a regional travel problem, an extension of the Multiple Traveling Salesman Problem (mTSP), to optimize the assignment and routing of responders—such as firefighters and volunteers—to affected individuals and facilities. To address the NP-hard nature of the problem, a computationally efficient heuristic is proposed that integrates fuzzy c-means clustering and a genetic algorithm (GA). Responders are first stochastically assigned to demanders based on a composite score combining distance, compatibility, and urgency. Remaining demanders are then optimally allocated using a GA to minimize total travel completion time while balancing workload. The model incorporates three key factors—workload differences, responder–demander compatibility, and urgency—and is implemented as a web-based travel assistance application capable of real-time recalculation when new responders or demanders appear.

Simulation experiments conducted in Setagaya Ward, Tokyo, demonstrated that accounting for workload differences and enabling dynamic recalculation significantly reduced completion time and improved cooperative task efficiency. Field experiments with actual responders verified these findings: the proposed system halved total completion time compared to conventional SNS-based coordination and eliminated route overlaps and missed visits. The results confirm that the proposed model and system enhance operational efficiency and reliability in dynamic disaster environments. This research provides a practical, data-driven foundation for real-time disaster management, with future work focusing on scalability, responder performance calibration, and robustness under disrupted network conditions.

Hyperparameter tuning is a critical step in training artificial intelligence (AI) models for Earth observation (EO) tasks, as it directly impacts model accuracy, convergence speed, and generalization capacity. Traditional optimization methods such as grid search, random search, and Bayesian optimization often suffer from high computational costs and limited scalability, particularly when applied to complex model architectures and large datasets. Grid and random search scale poorly with dimensionality of the search space and often waste evaluations on unpromising regions of the search space, especially for deep neural networks. Random search improves over grid search but still requires a large number of trials to reliably find good configurations in high-dimensional search spaces. Bayesian optimization methods, while more sample-efficient, typically involve non-trivial surrogate modelling and acquisition optimization steps that add overhead and can struggle with very large, mixed (discrete–continuous) search spaces. These challenges are further amplified in EO applications, where segmentation models are trained on large datasets, making each hyperparameter evaluation computationally expensive and limiting the practicality of purely classical search strategies.

Recent advances in quantum computing have introduced novel paradigms for solving combinatorial optimization problems. Quantum-inspired and hybrid quantum-classical algorithms leverage principles such as superposition and probabilistic amplitude encoding to enhance search efficiency in high-dimensional spaces while benefiting from the strengths of classical algorithms. Building on these concepts, we investigate a Hybrid Quantum Genetic Algorithm (HQGA) for hyperparameter tuning. To evaluate this approach, we apply it to the optimization of a semantic segmentation model specialized for wildfire burnscar detection.

This study presents a two-stage framework for planning Electric Vehicle (EV) Charging Infrastructure that explicitly targets Spatial Equity on Jeju Island. First, projected 2035 Origin–Destination demand is downscaled to a 500 m grid and evaluated on a routable road graph using a network-based Gaussian Two-Step Floating Catchment Area (G2SFCA) model to produce a high-resolution Accessibility surface. Second, a Quadratic Programming (QP) model jointly optimizes station Location and charger Capacity under a fixed budget by minimizing the demand-weighted variance of Accessibility, thereby reducing disparities across demand cells. Candidate stations are derived from publicly accessible Points of Interest and selected with a coverage-oriented clustering scheme; a greedy loop adds sites that yield the largest marginal reduction in the equity objective, with capacities re-optimized by QP at each step.

The evaluation compares four scenarios—status quo, Location-only, Capacity-only, and the proposed joint optimization—using established equity metrics including the demand-weighted Standard Deviation, Mean Absolute Deviation, Coefficient of Variation, and the Gini coefficient. Although full numerical results are in progress, preliminary simulations indicate that the joint strategy delivers more balanced Accessibility across urban and rural areas than single-focus baselines while maintaining overall service levels. The framework is reproducible, policy-oriented, and transferable to other regions, offering planners a rigorous, data-driven tool to allocate limited public fast-charging resources fairly under future EV uptake.

The “last meter” dilemma in urban accessibility refers to the lack of accessible information at the terminal points of public service facilities, which hinders wheelchair users' mobility, even physical infrastructure may be present. This study investigates this dilemma on a global scale, analyzing over 210,000 parks across 100 of the world's most populous cities to quantify how information gaps create real-world barriers.

To quantify these gaps, the study introduces two metrics: Absolute Accessibility Loss (AAL) and Accessibility Gap Ratio (AGR), which measure the additional travel time burden on wheelchair users resulted from the lack of accessible information. The findings show that only 34.9% of parks are labelled as accessible. This disparity has tangible consequences: Wheelchair users must travel farther and spend more time reaching parks labelled as accessible than the general population does to reach any park. The study also reveals a clear global divide, where high-income cities show higher labelling rates and shorter travel times for wheelchair users, while cities in Africa, India, and Southeast Asia exhibit higher disparities

This study furnishes a framework for policymakers, presenting a novel perspective for the assessment of urban equity and a scalable instrument for tracking advancements towards the United Nations Sustainable Development Goals, specifically SDG 11 (Sustainable Cities and Communities) and SDG 10 (Reduced Inequalities).

Image geo-localization aims to infer where a photograph was taken purely from its visual content. This task underpins applications in navigation, urban analytics, disaster response, and environmental monitoring, but current vision-language models (VLM) are mostly trained on generic web data with little explicit geospatial information. This work develops GeospatialCLIP, a geospatially enhanced VLM that embeds geospatial intelligence directly into CLIP via spatially explicit contrastive learning.

GeospatialCLIP is trained on 180k geotagged image-text pairs spanning street-view imagery, multi-temporal satellite images (2014 and 2023), and OpenStreetMap tiles. Rich captions and spatial context are curated by GPT-4 and experts, describing spatial patterns of objects, land use, urban form, and features that support geo-localization. A spatially explicit text encoder integrates structured tokens with geo-image type and geo-location across scales, enabling a shared geospatial representation space.

Zero-shot global geo-localization experiments evaluate GeospatialCLIP on unseen datasets across geo-locations, scales, and years, and compare it with vanilla CLIP and ResNet backbones. Across city, country and continent levels, GeospatialCLIP consistently improves top-1 accuracy for all imagery types, and its zero-shot performance on street-view images matches few-shot CLIP. The results highlight how embedding geospatial knowledge into VLMs can yield more robust, data-efficient GeoAI models and point towards future geospatial foundation models that better support scientific discovery and real-world decision-making.

Tourism and geographic text data is one of the most common data types in spatial analysis, and the classification of such data is an essential preprocessing step to facilitate more in-depth mining of spatial-temporal information. In the past decade, a variety of classification methods for tourism and geographical text data have been developed. These methods established important foundations for automated text analysis, yet their effectiveness has often been constrained by the availability of labelled data and the need for carefully designed feature representations. Recently, large language models (LLMs) demonstrate clear advantages in long-sequence modeling, offering new directions for text classification, particularly for long-form texts. However, employing commercial LLMs poses a significant cost challenge due to the high expense per token, and processing long texts consumes a considerable volume of tokens. In fact, it is feasible to adopt a strategy of locally deploying and fine-tuning open-source large language models that have reduced parameter counts. In this study, we have trained some open-source LLMs with chain-of-thought text. Experimental results show that the highest-performing model (e.g. fine-tuned Qwen3-1.7B) achieves an average accuracy of 95.83%, improving by 4.17% over the baseline RoBerta. Classification results can support tasks such as intelligent tourism recommendations, geographic knowledge construction, and toponym recognition. It may be concluded that the proposed chain-of-thought-guided LLM method can be effectively employed to classify tourism and geographic text data, and LLMs with reduced number of parameters have the potential to solve specific tasks with limited computation resources.

High spatio-temporal resolution estimation of XCO₂ is crucial for accurately quantifying regional carbon sources and sinks. Because XCO₂ variability is influenced not only by local geographic conditions but also by surrounding environmental and meteorological factors, this study proposes an advanced estimation approach that fuses multi-scale spatial features. We develop SpatialFusionNet, a convolution-based module that leverages local spatial association and receptive-field characteristics to integrate meteorological and surface environmental information within a 2.3° × 2.3° grid. This module extracts and fuses spatial feature patterns and subsequently estimates XCO₂ concentrations.

By combining SpatialFusionNet with machine learning methods—Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), and Deep Neural Networks (DNN)—we construct a deep spatial-feature fusion model based on OCO-2 XCO₂ observations over China, CAMS reanalysis data, meteorological variables, and vegetation indicators. Significant performance improvements are achieved: RMSE decreases by 1.297 ppm (SVM), 0.480 ppm (DNN), and 0.200 ppm (XGBoost) in ten-fold cross-validation against OCO-2 trajectory samples. Validation using the TCCON Hefei station yields a correlation of 0.85, demonstrating strong reliability.

Using the DNN combined with SpatialFusionNet, we further generate a seamless annual XCO₂ distribution for China in 2015 and analyze its temporal–spatial characteristics. The proposed framework provides an effective pathway for producing high-resolution XCO₂ datasets and supports fine-scale assessment of regional carbon cycling.

Urban strategies establishing climate shelters typically delineate service areas using 15-minute walking isochrones, aligning with "chrono-urbanism". However, this practice often relies on the standard walking speed of a healthy, middle-aged male, a simplification that risks significantly overestimating the real accessibility for vulnerable groups, such as the elderly.

This paper presents a dynamic methodology to analyse how accessibility changes when accounting for two crucial factors: age/gender and thermal comfort (heat exposure along the route). The approach uses the Physiologically Equivalent Temperature (PET) index to dynamically adjust walking speed based on environmental conditions and the heightened vulnerability of subjects (represented by a 65-year-old female).

Applied to a case study in Turin, Italy, the results demonstrate a profound accessibility error caused by standard methods. Neglecting the combined effects of age and heat may lead to a 100% overestimation of the actual number of elderly women served. When these factors were integrated, the municipal area covered by shelters plummeted from 35.2% (standard scenario) to only 8.6% (highest stress scenario). Furthermore, the proportion of elderly women considered served dropped drastically from approximately 65% to just over 18%.

These findings confirm that dynamic accessibility calculations are essential for identifying optimal locations for new climate shelters and ensuring effective, equitable adaptation strategies.

This study aimed to develop an easily accessible, interactive digital twin model in Unreal Engine that visualizes urban trees and their carbon flux based on the Metsäkanta tree database, and

simulated carbon sequestration and emissions dataset. The model provides a flexible and automated framework for incorporating additional carbon and tree data for any area. Additionally, it showcases the potential of data-driven game engine visualizations in creating engaging scientific communication for a broader demographic.

This contribution presents an exploratory study on the relationship between virtual and physical trajectories in London, with a particular focus on how their spatio-temporal lags evolve under different urban conditions. Virtual trajectories are derived from map tile access logs of OpenStreetMap, while physical trajectories are constructed from anonymised mobile phone data. Both datasets are aggregated to Middle Layer Super Output Areas (MSOAs) for the period from 1 February to 30 April 2020.

We apply a simple rolling-window cross-correlation to each MSOA to monitor, over time, whether virtual activity leads, coincides with or lags behind physical activity. Two case studies illustrate the insights provided by this spatio-temporal lag detection. Around major concerts at the O2 Arena, virtual trajectories consistently lead physical trajectories by approximately 1–3 days, reflecting anticipatory route planning and information searches. Around the first Covid-19 lockdown, the lag landscape reorganises: positive lags become more dominant and their spatial configuration shifts, indicating that virtual activity remains a robust leading signal for constrained but persistent urban mobility.

A Wikipedia-style collaborative mapping website is proposed in this paper to document, to archive, and to share these TEK. All knowledge articles are contributed by volunteers based on the volunteered geographic information (VGI) concept. The article can be written in the corresponding indigenous language to precisely describe their cultural knowledge. Compare to the Wikipedia, this website is actually a WebGIS. A knowledge article refers to a point, a polyline, or a polygon, which means the knowledge article is georeferenced. The website is composed of open software, such as MySQL, OpenLayers, GeoServer, Drupal and Apache. These indigenous knowledge articles are the source of contents of the Virtual Tribe, the virtual reality of their home tribe. We deployed UAV to take aerial photographs and produced ortho-rectified images and 3D mesh models of the tribe. We also applied 360°panorama camera to take 360°panorama images or videos at important locations when we walk with the elder people around the tribe. Finally, these images, 3D model, and TEK are integrated in the virtual tribe. It’s like a digital twin of the home tribe. Users can explore the tribe and learn TEK from elder people who speaks indigenous language in the 360° panorama video embedded in the virtual tribe.We have cooperated with two high schools in the indigenous countries to build up an immersive virtual reality (iVR) using the TEK articles on the proposed website. The feedback from students is positive and encouraging.

Locust outbreaks pose persistent challenges for agriculture and food security due to their pronounced spatiotemporal complexity. Existing monitoring and modelling approaches often struggle with sparse and biased field observations, cloud-affected and discontinuous satellite time series, and the difficulty of fusing heterogeneous data across scales. This paper reviews spatiotemporal locust modelling methods under a unified remote sensing–eco-statistical coupling framework. We first summarize multi-source observational inputs, including optical and microwave remote sensing, reanalysis meteorological data, and ground surveys, and outline common workflows for spatial alignment, temporal aggregation, lag handling, and uncertainty-aware quality control. We then examine three major model families along a coherent pathway from behavioural processes to probabilistic inference: Markov and semi-Markov models for explicit state transitions and duration; hidden Markov and state-space models for representing latent ecological states while correcting observation error; and hierarchical Bayesian spatiotemporal models, including INLA-based implementations, for cross-scale integration and formal uncertainty quantification. Building on this synthesis, we propose practice-oriented principles for model selection that account for state observability, temporal structure, spatial dependence, uncertainty representation, data and computational costs, and interpretability. Finally, we discuss a data–model–decision loop that links probabilistic risk products to operational thresholds, surveillance strategies, and control actions. The review aims to support more robust, transparent, and operationally useful early warning and resource allocation for locust management.

Recent increases in extreme climate events and urban accidents highlight the need for urban digital twin technologies capable of real-time monitoring and predictive simulation. However, existing digital twin systems primarily focus on visually realistic three-dimensional representations, which makes large-scale safety simulations computationally expensive due to massive 3D datasets and complex physical models.

To address this limitation, this study proposes a Hash interaction algorithm based on a variable 3D GeoHash framework for generating urban object information and enabling lightweight spatial interaction simulations. The framework extends conventional two-dimensional GeoHash by incorporating elevation to construct hierarchical 3D GeoHash cells that support efficient geocoding of urban objects.

The proposed method consists of four key processes: (1) classification of urban objects into fixed spatial information (e.g., buildings, roads, and terrain) and dynamic spatial information (e.g., weather conditions and moving entities); (2) generation of object-specific attribute information and physical properties; (3) establishment of movement rules between neighboring GeoHash cells; and (4) development of a rule-based inter-Hash interaction algorithm that updates physical state variables through interactions with adjacent cells.

By restricting interaction calculations to neighboring Hash cells, the proposed approach significantly reduces computational complexity while maintaining real-time update capability. The adjustable GeoHash resolution also enables simulations ranging from city-scale environments to centimeter-level spatial detail, supporting lightweight digital twin applications for urban safety management and construction-site monitoring.

This project proposes a GIS-based decision-support concept integrated within the electronic flight bag (EFB) to assist pilots in tactical in-flight weather deviation under convective conditions. The project addresses a critical gap between experience-driven cockpit decision-making, primarily relying on onboard weather radar imagery, and optimisation-based trajectory planning methods that are typically designed for strategic or air traffic management contexts rather than real-time pilot use.

The proposed framework utilises weather radar-aligned data combined with geospatial layers such as terrain, airways, and traffic to construct a unified operational environment. Within this GIS-based architecture, optimisation techniques (e.g., rapidly-exploring random trees, deep reinforcement learning) are applied to generate feasible and hazard-aware deviation trajectories. These trajectories are presented to pilots as advisory “ghost” flypaths on the EFB, supported by quantitative metrics such as weather clearance, additional track distance, and estimated fuel or time penalties, while maintaining the pilot fully in the decision loop.

Expected outcomes include improved flight efficiency through reductions in track mileage, deviation time, fuel consumption, and enhanced safety margins. Furthermore, the system aims to reduce pilot cognitive workload and stress by externalising complex decision-making processes and providing clear, optimised guidance during time-critical situations.

Overall, the project offers a practical cockpit-deployable solution that bridges weather radar-based situational awareness and advanced optimisation methods, enabling more consistent, data-driven, and operationally robust pilot decision-making.

This study proposes a Spatio-Temporal and Semantic Correlation Graph Convolutional Network (STS-GCN) to enhance the prediction of urban emergency events. Addressing the limitation of existing models that fail to fully integrate multi-dimensional correlations, the STS-GCN framework jointly models spatial, temporal, and semantic (categorical) dependencies. The model constructs distinct graphs to represent these relationships, using Graph Convolutional Networks (GCNs) to extract and fuse spatial and semantic features. A Gated Recurrent Unit (GRU) is then employed to capture temporal dynamics.

Trained and validated on a 2015 dataset from the Toronto Police Service—categorizing events into traffic collisions, shootings, robberies, and assaults—the model was evaluated against several baselines. Experimental results demonstrated that the STS-GCN model achieved superior performance, obtaining the lowest RMSE (0.1829) and MAE (0.0023), and the highest Accuracy (0.8705). The study concludes that through effectively learning the complex internal patterns of events through multi-dimensional feature modeling, the proposed framework offers a robust and generalizable tool for accurate urban emergency prediction, with significant potential to support public safety governance and resource allocation.

With the deep integration of geographic information science and human-computer interaction technology, how to support multiple users to cross different physical spaces and collaboratively perceive, analyze, and make decisions on complex geographic phenomena in a unified virtual and real fusion environment has become a cutting-edge challenge in this field. This article proposes a systematic mixed reality collaborative visual analysis method for the collaborative geographic cognition needs across geographic scenarios. The paper first analyzes the core scientific issues of cross geographical scenario collaborative analysis, namely the coupling representation of geographical scenarios and the collaborative aggregation of multi-user cognition. In response to this, we have constructed a four in one theoretical framework of "data model view interaction". The results show that this method can effectively break geographical isolation, build an immersive "co environment" collaborative space, and significantly improve the situational perception ability, communication efficiency, and collaborative decision-making quality of multi domain experts in complex geographical problems. This study not only provides cutting-edge collaborative analysis tools for geographic information science, but also provides important methodological support for interdisciplinary directions such as spatial human-computer interaction and group geographic cognition.

This study analyzes the spatial accessibility and distribution patterns of elderly care facilities in six central urban districts of Beijing (Dongcheng, Xicheng, Haidian, Chaoyang, Fengtai, Shijingshan) against the backdrop of rapid population aging. Using methods such as the Two-Step Floating Catchment Area (2SFCA) and kernel density analysis, the research integrates multi-source spatial and socio-economic data. Results reveal an unbalanced spatial distribution of facilities and varying service capacities, with insufficient coverage within a 5-minute travel scope and improved but transport-dependent accessibility within 10 minutes. The study highlights challenges in achieving “nearby elderly care”, particularly in areas like Fengtai District. It recommends optimizing facility layout by repurposing existing spaces in core areas and constructing new facilities in underserved peripheral zones, in line with “community-based” and “home-based” elderly care principles, to better meet the needs of the aging population.

As technologies such as intelligent connectivity and artificial intelligence become increasingly mature, and the platform economy evolves at a rapid pace, new products, business formats, and models—including autonomous driving, unmanned driving, and the low-altitude economy—are transitioning from pilot demonstrations to application trials, and beginning to enter widespread practical use on a large scale. The advancement of these new technologies, business formats, and models is driving the "ubiquitization" of surveying and mapping. It has become feasible to illegally obtain large-scale, precise location information of ground features quickly in a short period, which poses severe challenges to the supervision and administration of surveying, mapping, and geographic information security. This paper first introduces geographic information data security technologies, including the classification and grading of geographic information data and the confidentiality processing of geographic information data, among others. Secondly, it designs a geographic information security supervision and data governance model, covering geographic information data application scenarios, the geographic information data circulation control model, the geographic information data circulation security architecture and its applications, etc. Finally, it summarizes the challenges and opportunities faced by geographic information security supervision and data governance.

This study investigates the potential of GeoAI and instance-based segmentation for high-resolution land cover classification in San Vito di Cadore (Veneto), a UNESCO mountain region in the Dolomites characterized by high ecological heterogeneity. The dataset comprises 650 manually annotated orthophotos at a spatial resolution of 0.1–0.5 m, labelled across seven main land cover classes (Forest, Shrubland, Grassland, Cropland, Water Bodies, Artificial/Urban Areas, and Rocky/Bare Areas) and harmonized with Corine Land Cover (CLC) aggregated categories for inter-comparison. Snow and cloud were treated as auxiliary classes given their frequent occurrence in alpine imagery. The YOLOv11 instance-segmentation model was trained on 1000×1000 px tiles, with a SAHI (Slicing Aided Hyper Inference) framework adopted during inference to process large-scale orthophotos without loss of spatial quality. Results show an overall precision of 0.847 and recall of 0.575, with mAP@0.5 exceeding 0.65. Quantitative comparison with the Regione Veneto land cover product (2023) reveals good agreement for the dominant forest class (−1.4%), while the largest discrepancies concern artificial surfaces (+20.8%) and agricultural areas (−36.3%), attributed to differences in spatial scale and training-sample imbalance. The work highlights the advantages of instance-aware deep learning for generating accurate, spatially coherent land cover maps and underlines the growing relevance of GeoAI workflows for environmental monitoring and spatial planning in complex mountain environments.

Urban building clusters can significantly alter the low-altitude urban wind field, creating local regions with high wind speed, intense turbulence, and strong non-uniform disturbances. These conditions may cause lateral deviation, attitude instability, increased trajectory error, and even exceed the wind-resistance limits of low-altitude UAVs. To address the lack of explicit modeling of local high-risk wind regions in traditional path planning methods, this paper proposes a wind-risk-aware path planning method, termed RP- A* (Wind Risk-Aware Probability A*). First, a steady-state RANS urban airflow model with the standard k-ε turbulence model is used to obtain the mean wind field and turbulence statistics in the low-altitude flight region. Second, the CFD results are reconstructed on a grid to build a wind risk model consisting of gust risk, crosswind risk, and overall wind-limit-exceedance risk. Third, a direction-dependent integrated risk cost is introduced into the A* search framework, and the RP-A* algorithm is developed to achieve path planning that balances route efficiency and flight safety. Finally, Monte Carlo simulations driven by turbulence-based wind perturbation samples are conducted to estimate the empirical failure rate of planned paths. Results show that RP-A* significantly reduces path failure risk compared with baseline shortest-path methods while requiring only a limited increase in path length. The proposed framework provides an effective approach for safe UAV path planning in complex urban low-altitude environments.

The sustainable development of historic districts, aligned with UN SDG 11.4, requires integrated approaches that balance heritage preservation with contemporary urban functionality. This study proposes a novel analytical framework combining Space Syntax theory and Point of Interest (POI) data to address this challenge. Departing from traditional non-hierarchical methods, approach of the study innovatively processes vectorized road networks with a focus on community and block-scale hierarchy, more accurately reflecting human-scale spatial perception and connectivity. This refined Space Syntax model quantitatively analyzes street accessibility and spatial configuration, which is then integrated with categorized POI data to reveal the inherent socio-economic logic and functional distribution within historic urban areas. The framework is empirically validated through a case study of a historic district in Kunming, China. Results demonstrate that this combined methodology provides a comprehensive understanding of the spatial organization, offering data-driven insights to support precise and sustainable conservation and renewal strategies for historic districts.

This study addresses the challenge of automatically recognizing and extracting spatial coordinates from unstructured natural language texts, particularly within the domain of land registry documents (fiduciary documents). It proposes a deep learning-based method utilizing the BERTimbau language model, fine-tuned for the Named Entity Recognition (NER) task. This research expands the scope of geoparsing beyond the traditional focus on toponyms, specifically targeting the direct extraction of coordinate data for reliable automation in engineering, land cadastre, and land regularization.

Urban densification in post-socialist cities involves fine-scale spatial transformations that are difficult to quantify in data-scarce environments.

This study proposes FLAIR-HUB2BF, a model inference-based workflow for automated building footprint extraction

and multi-temporal change analysis, applied to the city of Iasi, Romania. The methodology extends the SUBDENSE conceptual

framework by integrating the FLAIR-HUB deep learning model for semantic segmentation of very high resolution aerial orthophotos

from 2011 and 2024, followed by binary mask extraction, instance segmentation, and Douglas–Peucker polygon generalization,

yielding 34,454 and 17,141 georeferenced building footprints, respectively. The approach demonstrates that coherent building

footprint datasets and their temporal evolution can be derived directly from aerial imagery without relying on complete cadastral

databases. To support rigorous evaluation, the first open benchmark building footprint datasets for Romania were produced through

manual photo-interpretation correction across four morphologically distinct urban neighborhoods of Iasi, and assessed against ISO standards

19157 spatial data quality standards, achieving commission and omission rates of 1.95% and 2.39%, respectively. Quantitative

evaluation using complementary GMA (Geometric Matching of areas) and MCA (Multi-Criteria Algorithm) data matching algorithms

confirms moderate-to-high spatial accuracy, with MCA surface agreement rates reaching 91%. The results demonstrate

the capability of the method to capture fine-scale urban transformations, including infill development, brownfield redevelopment,

and peri-urban expansion, while revealing the critical influence of input data quality on segmentation performance. The proposed

workflow establishes a transferable, reproducible, and open methodology for building-level urban monitoring applicable to other

Romanian and European cities facing similar data constraints.

Augmented Reality (AR) offers an immersive medium for visualizing 3D city models directly within physical environments, but current systems lack real-time synchronization with authoritative geospatial databases. This paper suggests an open-source architecture that bridges this gap by enabling bidirectional, standards-compliant communication between AR Microsoft HoloLens 2 frontend and a CityGML-based backend. The system integrates PostgreSQL/PostGIS with 3DCityDB, exposed through a Django API, and connects to AR front-ends such as Microsoft HoloLens 2 via Cesium for Unreal.

This work proposes an extension to the OpenDRIVE standard to represent pavement surface defects, improving the realism of autonomous vehicle simulations and enabling the integration of road condition data from modern mapping and AI-based detection methods.

In cultural heritage research, one task is to reconstruct historical ship routes; however often exact trajectories are lacking, and thus an accurate itinerary is difficult to reconstruct due to inaccurate or incomplete documentation. The aim of this work is therefore to create spaces of uncertainty as so-called possibility spaces based on the calculation of a geographical extent that attempts to encompass all valid path options.

In order to obtain meaningful possibility spaces, it is important to first define the navigable space and also take into account additional factors such as water depths, currents, wind direction and known historical shipping lanes that may influence a possible route. This information can be used to define the cost for calculating possible routes.

To calculate possibility spaces, different approaches of path planning are explored, such as transferring the navigable space into a routing graph, converting the space into a regular grid, or using an irregular grid in terms of a mesh.

Subsequently, options for deriving the final possibility spaces are described, such as using the explored nodes during the search process (e.g. using A*), or to generate a possibility space by creating a variation of paths by calculating k-shortest paths. Of particular interest is the calculation of paths that have a cost value similar to a predefined acceptable maximum. These paths form the outer boundaries of the possibility space to be created.

High-definition (HD) maps are a key digital infrastructure for connected and autonomous vehicles, especially in highway environments where detailed and reliable road representations are required. This contribution presents an end-to-end workflow for HD road map generation from mobile mapping data, developed within the HD SMART MAP project (PNRR Spoke 7) and applied to a 10 km stretch of the Tangenziale di Napoli.

The survey was carried out with the GAIA M1 Mobile Mapping System, integrating LiDAR, panoramic imagery and GNSS/INS. This configuration enabled the acquisition of dense point clouds and synchronized images even in GNSS-challenging areas such as tunnels. All data were georeferenced in the national reference system RDN2008, with heights referred to the ITALGEO2005 geoid.

The processing pipeline includes point cloud filtering, ground segmentation and DTM generation, as well as the production of an orthophoto of the corridor from panoramic imagery. These products support the semi-automatic extraction of lane markings and road features, which are then encoded according to the ASAM OpenDRIVE standard.

The resulting HD map provides a geometrically and semantically rich, machine-readable description of the highway, suitable for vehicle localization, path planning and simulation. The case study demonstrates that semi-automated procedures significantly speed up HD map production compared to traditional manual workflows.

Vector geometries are traditionally represented as entities of sequences of coordinate structures. With advancements in hardware and software for data analytics, a preference for columnar data layouts arose. This paper examines array programming for spatial operations to evaluate the potential performance benefits of modern computing architectures and emerging spatial data encodings. Evaluating selected operations, such as distance calculation, extent extraction, and affine transformations, indicates similar or improved performance for geometries with columnar coordinate layouts. By leveraging modern compiler infrastructure, we further demonstrate that advanced hardware features in commodity hardware, such as vector instructions, are becoming accessible without specialized code. The performance comparison with established, widely used geospatial and computational libraries reveals significant untapped potential for increasing the efficiency of spatial computing.

Recent advances in geomatics and 3D surveying have enabled the acquisition of increasingly dense point clouds through both static and mobile laser scanning systems, supporting the rapid digital documentation of built and natural environments. At the same time, the growing diffusion of WebGL technologies has opened new possibilities for the online visualization and dissemination of complex three-dimensional datasets. Within this context, the present study proposes an open-source workflow for the automatic extraction of significant geometric surfaces from laser scanner point clouds and their integration into a web-based visualization framework. The method was developed within a Python-based processing environment and tested on three datasets characterized by different levels of geometric complexity: a regular built environment, an under-construction building environment, and a historical context. The workflow includes point cloud preprocessing, automated segmentation strategies adapted to the geometric complexity of each case, extraction of planar and non-planar elements, polygonal surface generation, mesh construction, and conversion of outputs into lightweight formats suitable for web publication. The final visualization environment combines segmented polygonal models and subsampled point cloud data through open-source WebGL technologies. The results demonstrate that the proposed strategy provides a scalable and flexible solution for the rapid online representation of large laser scanner datasets, supporting surface recognition, low-cost accessibility, and future semantic enrichment within web-based geospatial and Digital Twin applications.

This paper reports progress in the Venaria Reale pilot of the EU Horizon Europe project HERITALISE (2025–2028), using the Galleria Grande as a test case for a preventive-conservation workflow toward digital twin integration. It presents a reproducible Scan-to-BIM workflow for HBIM that delivers a 3D backbone combining geometric reliability, semantic queryability, and source traceability. Multi-sensor datasets, including terrestrial laser scanning, SLAM-based mobile mapping, and UAV photogrammetry, are georeferenced within a unified coordinate framework. A georeferenced UAV–TLS fused point cloud serves as the main modelling baseline, while SLAM data are used only as local infill for verified missing areas. Data management follows a raw-working-deliverable structure with logged parameters, transformation matrices, and quality-control notes. Registration residuals are controlled within 0.01–0.05 m and checked through section-based and distance-based validation in critical junction areas. Geometric modelling adopts a Revit-Rhino workflow guided by structural, semantic, evidential, and feasibility criteria. Semantic enrichment follows the HERITALISE Common Data Environment and BIM Execution Plan, with ObjectID linking HBIM elements to an external SQL database and supporting continuity between legacy and current room naming systems.

Landslide susceptibility assessments have traditionally used static rainfall statistics that do not reflect the actual meteorological conditions when slopes fail. This study develops a machine learning framework that aligns high-resolution radar rainfall (XRAIN, 250 m / 10 min) and modeled soil moisture (XSWI) with documented landslide occurrence times as dynamic triggering factors. Applied to the Heavy Rain Event of July 2018 in Hiroshima Prefecture, the framework combines watershed-based spatial cross-validation, systematic comparison of four class imbalance strategies (no treatment, sample weighting, random under-sampling, and SMOTE) across eight algorithms (XGBoost, LightGBM, CatBoost, HGBoost, Random Forest, Balanced Random Forest, Easy Ensemble Classifier, and Logistic Regression), and spatially explicit SHAP interpretation.

Two key findings emerged. First, soil moisture state — not instantaneous rainfall intensity — was the dominant triggering predictor: XSWI variables ranked 2nd and 3rd in importance after slope angle, operating as independent axes (r = 0.074). The no-treatment condition consistently outperformed all resampling strategies across algorithms. Second, spatial SHAP mapping revealed that predisposing factors produce time-invariant contribution patterns governed by terrain, while dynamic triggers produce event-specific patterns reflecting rainfall distribution; their spatial overlap identifies the highest-risk locations. Time-series susceptibility maps confirmed that the framework captures within-event risk evolution as rainfall progresses — a capability unattainable with static approaches. These results indicate that incorporating occurrence-time-aligned soil moisture dynamics substantially improves both the predictive and explanatory capacity of landslide susceptibility assessment.

Accurate air quality monitoring depends on continuous satellite observations of key pollutants such as nitrogen dioxide (NO₂) and sulfur dioxide (SO₂) from the Sentinel-5P/TROPOMI mission. However, these datasets often suffer from severe spatio-temporal discontinuities due to cloud cover, surface reflectance, viewing geometry, and strict quality filtering, which limit their reliability for environmental and health-related applications.

This study addresses the challenge of missing data reconstruction over the Po Valley, Northern Italy (2019-2023), an area characterized by complex terrain and frequent winter inversions that amplify data gaps. A comprehensive statistical analysis revealed substantial data loss, averaging 45% for NO₂ and 77% for SO₂, with pronounced seasonality and strong correlations with elevation.

To fill these gaps, an integrated machine learning framework was developed, combining a LightGBM baseline model and a 3D Convolutional Neural Network (3D CNN). The models exploit multi-source predictors, including meteorological variables (ERA5), atmospheric priors (CAMS), topography, land cover, and population density, together with cyclic temporal encoding.

Preliminary results demonstrate that the 3D CNN significantly improves gap reconstruction performance (R² = 0.95 for NO₂, 0.74 for SO₂) compared to the LightGBM baseline, though at higher computational cost. The proposed framework enhances the spatio-temporal continuity and usability of Sentinel-5P data, supporting more reliable environmental monitoring and policy-making in data-sparse conditions. Future work will extend the approach to other pollutants, regions, and deep learning architectures.

The Ecuadorian Amazon hosts remarkable biodiversity and cultural richness but faces increasing pressures from the expansion of oil-related activities. This study evaluated the distribution and concentration of deforestation hotspots between 2000 and 2023, analyzing their relationship with existing oil infrastructure and environmentally significant areasLand Use and Land Cover data from MapBiomas Ecuador were combined with Kernel Density Estimation (KDE) analyses based on the spatial distribution of oil blocks, pipelines, wells, and the limits of environmentally sensitive areas. The results indicate a net loss of 391,303 ha (4%) of forest cover, with 80% of the hotspots located within a 10 km radius of hydrocarbon infrastructure. However, intangible zones, protected areas, and water protection zones showed minimal impacts. The findings of this study provide technical evidence to support land-use management and conservation efforts in ecologically vulnerable Amazonian regions.

Nighttime light (NTL) Earth observation data represent an invaluable resource for measuring population distribution, disaster impact, economic activity, and socio-economic inequalities from space. While traditional NTL data sources provide consistent long-term measurements, they are spatially coarse, impeding spatially detailed analyses of nighttime lights. Novel, high-resolution NTL data from the SDGSAT-1 satellite capture NTL variations across space and time at fine spatial detail of 10 to 40 m and open new research avenues but also require innovative analytical approaches. Herein, we demonstrate the potential of jointly assessing annual SDGSAT-1 composites and human settlement data from the Global Human Settlement Layer (GHSL) and other data, characterizing the built environment, human population distribution, and the rural-urban continuum. We illustrate that such data integration generates new insights on the interactions of nighttime lights and settlement-related characteristics at unprecedented detail. For example, we find that NTL emissions tend to be highest in old parts of settlements (<1975) and lowest in very recently developed land. The brightness of major roads and non-residential areas at night approximately doubles, on average, compared to residential built-up areas. ~80% of urban population resides in areas characterized by luminous, stationary NTL, while this population share drops to ~15% in very rural areas. Looking at infrastructure-related land use, we find that airports emit the highest levels of stationary and non-stationary NTL in our study area. These results illustrate the potential of high-resolution data from SDGSAT-1 and pave the way forward to include such data in settlement and population modelling at scale.

The POSEIDON project aims to develop scalable and repeatable approaches for monitoring Posidonia Oceanica (PO) meadows, a key Mediterranean habitat that supports coastal ecosystem services and long-term blue-carbon storage, yet they are increasingly threatened by warming waters and cumulative human pressures. This work presents a satellite-based workflow for benthic habitat mapping that combines Sentinel-2 multispectral imagery, ancillary bathymetry, and deep-learning semantic segmentation. Sentinel-2 Level-2A data and bathymetry were integrated into multi-band inputs on a common 10 m grid, with analysis restricted to water pixels. A wall-to-wall reference map was generated by harmonising existing habitat products into six benthic classes for supervised model training and evaluation. U-Net and DeepLabv3 architectures with a ResNet backbone were tested for a representative September 2015 scene. The workflow was first assessed in the Culuccia peninsula, where it achieved an overall accuracy of 0.830 and a Kappa coefficient of 0.786. It was then successfully transferred to the Capo Testa - Punta Falcone Marine Protected Area (MPA), where the best-performing configuration reached an overall accuracy of 0.882 and a Kappa coefficient of 0.843. These results show that open-access satellite data combined with robust semantic segmentation models can provide a reliable and non-destructive framework for seagrass mapping in complex coastal environments.

Nowadays, with technological advancements and the increasing volume of urban traffic, low-altitude urban air mobility, particularly for time-sensitive trips such as airport travel, has emerged as a promising solution. Vertiports are one of the key components of this novel transportation system, serving as the ground connection points for urban air mobility operations. The optimal placement of vertiports, considering various influencing factors, is critical to the successful implementation of this emerging mode of transportation.

In this study, a data-driven framework is proposed to identify the most suitable areas for vertiport placement in order to facilitate and accelerate access to airports in the City of Toronto. By integrating environmental constraints, population density, ground transportation connectivity, noise impact zones, and regulatory considerations, the framework evaluates land suitability using GIS-based analysis and a deep-learning approach called Generative Adversarial Network (GAN). The proposed methodology can generate a vertiport network by learning nonlinear spatial relationships between multiple spatial layers, without the need for subjective rules, and finally identifying potential vertiport locations with maximum coverage. The results demonstrate two strategically located vertiports for accessing each of Billy Bishop and Pearson airports, situated in commercial, mixed-use, and industrial zones, high-demand areas, and locations near major public transit stations.

In this paper, we investigate the use of CityGML data for visual localization. Therefore, we present a visual localization approach that uses CityGML data. We compare different matching approaches for real images and renderings of CityGML data and evaluate our results using query images with accurate ground truth poses. We show that pose estimation is possible with object features of city models. We propose an evaluation of the estimated pose with independent ground truth poses from the reference data.

Achieving a genuine digital twin for smart buildings requires accurate and real-time knowledge of the spatial positions of users and objects within indoor environments. Despite major advances in Indoor Positioning Systems (IPS), most existing frameworks lack a structured data connection with Building Information Modeling (BIM) which provides the semantic and geometric representation of building elements and spaces. This disconnects limits both real-time synchronization and three-dimensional spatial visualization. This study presents a novel BIM-driven hybrid framework that integrates Wi-Fi and Bluetooth Low Energy (BLE) signal data with BIM to establish a data-centric foundation for digital-twin development. The experimental setup was deployed on the fourth floor of the Faculty of Geography at the University of Tehran, modeled as a BIM-based indoor test environment. Received Signal Strength (RSS) data were collected from 35 reference points (RPs) and seven transmitters (four Wi-Fi access points and three BLE beacons), normalized, and processed using both Fingerprinting and Trilateration models. By incorporating the vertical component (Z) and linking spatial records to BIM entities such as IfcStorey and IfcSpace, user locations were visualized within a 3D building model. The BLE-enhanced Fingerprinting model achieved a substantially lower error (RMSE ≈ 0.40 m) than Trilateration (RMSE = 2.38 m), while the final hybrid model, employing adaptive weighting between sub-models, achieved more than 95 % accuracy within one meter of the ground truth.

These results demonstrate that integrating IPS data with BIM provides a robust foundation for digital-twin creation in smart buildings.

Bottom-up, physics-based Urbain Building Energy Modelling (UBEM) approaches enable systematic assessment of building typologies and operational behaviours even when empirical data are limited, providing robust results for district-scale heating and cooling simulations. However, most physics-based UBEM applications have focused chiefly on building-related parameters and have given limited attention to environmental factors such as vegetation, although this element affects building energy demand under changing climate conditions.

To overcome this limitation, the paper will present a simulation-based workflow that evaluates how urban tree-planting strategies influence building heating and cooling demand under current and projected 2050 climate conditions. Specifically, the workflow builds upon the simulation-based UBEM platform SimStadt by embedding vegetation effects directly within a single modelling environment, removing the need for external microclimate coupling or additional simulation tools. Our method is based on standardised CityGML building models, simplified yet seasonally dynamic vegetation representations, and a unified modelling environment that allows consistent comparison of cooling and heating demand under both current and projected climate conditions. This integration allows for the quantitative evaluation of tree-planting strategies for both heating and cooling demand at the district scale.

The paper will present the results of the study carried out in some neighbours of the Dutch city of Rotterdam.

Cities are at the core of the current debate on climate change mitigation, and multiple policies on the global and continental scale have acknowledged this condition, pushing towards an increase in the renovation rate and the installation of renewable energy technologies. The revised Energy Performance of Buildings Directive (European Parliament and Council, 2024) targets the renovation of 35 million buildings across Europe, starting from worst-performing buildings. The authoritative tool for the identification of such buildings is the Energy Performance Certificate, the European reference scheme for energy performance in buildings, which covers only a fraction of the European building stock, approximately 30-50%. This study aims to refine a published methodology which takes advantage of aerial thermography and Energy Performance Certificates to attribute an energy class to the whole building stock. The research question is how to classify the building stock, thus making it possible to compute the final energy consumption, by adopting a geospatial approach which considers simultaneously remotely-sensed and metered data. This research considers three main inputs: a thermal point cloud, the building layer of the technical map of the City of Turin, and Energy Performance Certificates.The method is based on the assumption that all buildings have the same indoor temperature. For this reason, the external surface temperature is a proxy of the quality of the envelope, with low-performing buildings having higher thermal losses and therefore an higher external temperature. Then, the class distribution is observed in the Energy Performance Certificates database and replicated in the whole building stock.

Time series are an important source of information about changes in land cover. However, historical raster datasets are often characterized by vague and imprecise temporal properties. We have developed a novel raster data management system designed specifically for machine-learning applications, which organizes temporally vague raster data in cloud environments. The system addresses the challenges of processing historical maps with uncertain temporal attributes. It combines object storage, PostGIS Raster and the Spatio-Temporal Asset Catalogue (STAC) API, enabling the efficient, interoperable management of spatio-temporal raster data. It allows users to define and evaluate vague instants and fuzzy intervals, enabling them to perform precise queries on temporally relevant datasets. This solution is particularly useful for managing databases in a flexible and customizable way, and is ideal for sovereign data management and self-managed infrastructures.

This study investigates the long-term evolution of Kenya’s road network from 1950 to 2020, highlighting how colonial legacies, post-independence modernization, and contemporary planning have shaped infrastructure development. Using deep learning techniques, roads were extracted from historical topographic maps (1950–1980) and transformed into GIS-compatible data, resulting in a nationwide road dataset of approximately 56,000 km from the mid-20th century. These data were compared with a 2020 dataset from the Kenya Roads Board, which documents over 161,000 km of roads. The analysis reveals that Kenya’s total road length has nearly tripled, and the average distance to the nearest road has decreased from 8.6 km to 5.1 km. However, the road development is uneven across the country: southern and central regions show significant growth, while northern and arid areas remain underserved, reflecting persistent spatial disparities rooted in colonial planning. A regional comparison in southwestern Kenya shows a 56% increase in road length between the 1950s/60s and 1970s/80s, with notable upgrades in road quality. The proportion of paved roads rose from 1.5% to 12%, and tertiary dry-weather roads declined from 64% to 26%. Despite these improvements, only 15–30% of Kenya’s roads are paved today, which is below the continental average of 47%. This study demonstrates the value of integrating historical and contemporary geospatial data to assess infrastructure development, identify gaps, and support planning aligned with Kenya Vision 2030 and the Sustainable Development Goals. The findings underscore the importance of spatial analysis in evaluating development outcomes and guiding future investment strategies.

Knowledge of the regional distribution of the world’s population is essential for political and social decisions not only but especially for achieving the 17 Sustainable Development Goals (SDGs). Census and other population data at the subnational level are important for this purpose. However, current population data management platforms largely ignore the spatiotemporal nature of census data. Here, we outline the requirements for a spatiotemporal population data management system and present its general architecture, data model and state of implementation. The developed system currently stores population data from approximately 200 countries, nearly 11 million spatial units and around 770 million individual population figures.

Managing dynamic geographic knowledge effectively is hindered by fragmented tools lacking holistic integration, particularly when handling the heterogeneous and evolving nature of real world spatio-temporal data. Traditional knowledge graphs and databases struggle with efficient representation, storage, and reasoning over such complex information. This paper propose an integrated computation framework built upon Grid Graph Modelling to make geography computable. This framework provides an end-to-end solution encompassing knowledge representation, storage, querying, and spatio-temporal reasoning. It synergistically integrates three core components: the Grid Augmented Geographic Knowledge Graph (AugGKG) for unified grid based representation with computable spatial relations; the Grid Graph Database (GGD) for spatially aware storage and efficient grid algebra based computation; and the Grid Neighborhood-based Graph Convolutional Network (GN-GCN) for advanced reasoning by learning from semantic, spatial grid, and temporal dimensions. This cohesive architecture transforms diverse geographic data into actionable knowledge, enabling efficient querying and complex reasoning, paving the way for next generation intelligent geospatial systems, including empowering foundation models, enhancing smart cities, creating digital twins, and reasoning geographic event evolution.

This paper introduces a method to analyze the effect of aggregation on continuous (interval or ratio scale) raster data. Previous research used the entropy based local indicator of spatial association (ELSA) to study the change in the local spatial association this, new paper extends that idea by evaluating both the within and between pixel variability. The standard deviation was used to evaluate the between pixel variability with a decrease in the SD indicating a decrease in the image information content. Ec (diversity) is one part of the ELSA statistic and gives a measure of the within-pixel heterogeneity. We should balance the this within and between-pixel variability when choosing the pixel size for a raster dataset. The variogram was used to explore the change in spatial structure. Current research is refining this method and developing a tool that will support the user to choose the pixel size for mapping.

Current research is following two further avenues. The first is to adapt this method for categorical data with an application in land cover mapping. Second is to build in the effect of predictive uncertainty in the pixel values.

Carrier phase observations enable millimeter-level GNSS positioning, but their continuity is frequently disrupted by signal blockages and cycle slips. This limitation is particularly critical for low-cost and smartphone receivers, where weak antennas, urban multipath, and duty cycling cause frequent phase gaps that prevent reliable ambiguity resolution. Before addressing the full complexity of mass-market observations, the prediction methodology must be validated under controlled conditions. In this work we investigate whether machine learning, supported by precise satellite orbits and clocks, can predict carrier phase observations during signal gaps with millimeter-level accuracy. Twenty-four hours of Galileo data from the TORI permanent station (SPIN3 network, Torino, Italy) are processed at 30~s sampling using GFZ final SP3 and CLK products. After forming the ionosphere-free combination, an iterative carrier-phase based estimator removes the receiver clock, tropospheric delay, and ambiguity, reducing the residuals to a median standard deviation of 60~mm. Synthetic gaps from 60~s to 1800~s are introduced (1045 gaps total) and four prediction strategies are compared: polynomial fitting (degrees~3 and~5), Fourier-augmented polynomial, Gradient Boosting Regression with satellite geometry features, and Gaussian Process Regression. The Gradient Boosting model achieves the best overall performance, reaching 4.4~mm RMS for 60~s gaps, 9.4~mm for 5~min gaps, and 21~mm for 30~min gaps, well below the half-wavelength threshold required for cycle slip repair. These results demonstrate that geometry-aware gap prediction is feasible at the sub-wavelength level, providing a validated foundation for extending the approach to low-cost and smartphone GNSS receivers.

Indoor maps are essential for robot services, but the high dynamics of indoor environments caused by human activities lead to frequent layout changes, making it challenging to maintain map accuracy and timeliness. Existing map update methods, such as periodic full reconstruction or event-triggered incremental updates (Prieto-Fernández et al., 2024; Xia et al., 2024), lack a quantitative mechanism to evaluate whether element layouts are sensible. This makes it difficult to predict systematic changes and creates a paradox between "update frequency and element granularity." To overcome these limitations, this study proposes a spatial cognition-driven approach to identify the rationality of indoor element layouts, providing a predictive metric for efficient, layered map updates and enabling advanced robot navigation and safety warnings.

To address limited high-level semantic modeling in indoor 3D point cloud segmentation, this study proposes a text-knowledge-guided framework built upon RandLA-Net. Category-level textual semantic prototypes are constructed through multi-template prompting and encoded by a pre-trained text encoder to provide stable semantic priors. These textual cues are progressively integrated into point cloud feature learning through shallow semantic modulation and high-level cross-modal fusion, enhancing the interaction between geometric representations and semantic knowledge. The network is jointly optimized by segmentation supervision, prototype alignment, and boundary refinement, enabling it to learn discriminative features that preserve local geometric details while encoding richer semantics.

Superpoint matching is a critical step in coarse-to-fine point cloud registration, and its performance directly affects the accuracy of sub-sequent point matching and pose estimation. However, most existing methods establish correspondences mainly relying on feature sim-ilarity, without explicit modeling of spatial structure, which easily leads to unstable matching in complex scenarios such as noise, occlu-sion, and low overlap. To address these issues, this paper proposes a coarse-to-fine point cloud registration method guided by prior correspondences. First, prior superpoint correspondences are constructed using rigid transformations estimated by existing SOTA methods, and are serially encoded via a prior encoding module to provide explicit constraints for feature learning. Furthermore, multiple geometric information including pairwise distances, angles, and normals is introduced and uniformly encoded to enhance spatial struc-ture representation. On this basis, a prior-guided sparse mixture-of-experts attention mechanism is designed to differentially model fea-tures in overlapping and non-overlapping regions, thereby improving feature discriminability and structural consistency. Using the learned features, the model gradually establishes correspondences through superpoint matching and point matching, and estimates the final rigid transformation with RANSAC. Experiments on the 3DMatch dataset show that when sampling 1000 point correspondences, the proposed method achieves an inlier ratio of 80.7% and a registration recall of 92.9%, which are 5.5% and 1.1% higher than the base-line method respectively, verifying the effectiveness of the proposed method in terms of accuracy and robustness.

Climate change is reshaping the global water cycle, causing substantial alterations in precipitation, evaporation, and runoff patterns. These shifts are driving rapid changes in terrestrial water storage (TWS), which includes groundwater, soil moisture, surface water, snow, and ice. Declining TWS threatens freshwater security, increases the vulnerability of ecosystems and communities, and directly impacts sustainable water management—key concerns addressed under SDG 6 . In parallel, intensifying climate-driven water losses align with the global challenges highlighted in SDG 13, particularly regarding adaptation and resilience.

This study examines long-term TWS variations in the Susurluk Basin of Türkiye’s Marmara Region using NASA’s GRACE and GRACE-FO satellite missions. By measuring gravity anomalies caused by mass changes, GRACE enables the detection of large-scale water storage shifts. Monthly data from 41 GRACE grid points (2002–2022) were processed using the Mann-Kendall trend test at a 5% significance level. Consistent acceptance of the H1 hypothesis and universally negative Z values confirm a statistically significant and persistent decline in TWS across the basin.

Results show that water storage loss accelerated dramatically between 2012 and 2022 compared to 2002–2012. The basin exhibits an overall decreasing coefficient of –0.0561, while sub-basin analyses indicate 20-year average losses ranging from –1.3 cm to –0.1 cm. These findings demonstrate a clear, worsening depletion of water resources, emphasizing the urgent need for climate-adaptive water management. The documented TWS decline underscores the relevance of this work to SDG 6 by highlighting risks to water availability and to SDG 13 through evidence of climate-induced hydrological change.

The aim of this paper is to design and develop an openly accessible, web-based Crowdsourced Geographic Information (CGI) framework, referred to as the Digital Detectives of Environment (DiDE), to facilitate the collection of geo-located events. The framework incorporates three user roles: (i) citizens, (ii) experts, and (iii) supervisors. Citizens can browse relevant events without requiring authentication, while experts are responsible for collecting geographic data, including the optional attachment of photographs or videos. Supervisors, on the other hand, define and manage event types. Each event type is classified as either useful or harmful, which determines its visibility to citizens. The pilot implementation was conducted at the Beytepe Campus of Hacettepe University, focusing on four event types aligned with Green Deal actions: rubbish bins and recycling bins (useful), and cigarette butts and full rubbish/recycling bins (harmful). During a one-week data collection period, a total of 490 events were recorded by 37 students. The results reveal clear clustering patterns in both space and time. Temporally, a large proportion of the data was collected on the final day, indicating a tendency toward procrastination among participants. Spatially, events are concentrated in the southern part of the campus, where most facilities are located. This pattern is further supported by analyses using the F and G functions. In particular, cigarette butt events exhibit strong spatial clustering, with a mean nearest-neighbour distance of approximately 25 metres. This finding provides empirical support for the broken windows theory.

Pavement Management Systems (PMS) are essential for evaluating and maintaining transportation infrastructure; however, conventional monitoring methods are often labour-intensive, costly, and inaccurate. The growing need for reliable. timely pavement condition data has driven the development of automated, data-driven approaches. This study presents a low-cost and scalable framework for pavement condition monitoring that integrates multimodal sensing with a digital twin (DT) environment. Smartphones equipped with inertial measurement unit (IMU) sensors, GPS, and cameras are used to collect synchronized vibration and visual data during normal driving conditions. Vibration signals are analysed to detect anomalies associated with pavement surface irregularities, while video data are processed using a deep learning-based object detection model to identify surface distress. A late fusion approach combines the outputs from both modalities to improve detection reliability and provide comprehensive condition assessment. The system enables spatial mapping of detected distresses and supports real-time visualization through a web-based DT dashboard. Results demonstrate that multimodal sensing compensates for the limitations of individual sensors, enhancing both detection accuracy and robustness. The proposed framework offers a practical solution for efficient pavement monitoring. It supports data-driven decision-making for proactive infrastructure management, with potential for future expansion through crowdsourced data and additional sensing technologies.

To address the challenges of low efficiency and omission in manual inspection of small road debris, this study develops a lightweight mobile monitoring system for fine road debris. The system integrates a high-resolution industrial camera and a GNSS positioning unit to achieve real-time image acquisition and spatial synchronization. Built on the Python platform, the software includes data acquisition and communication modules that enable automatic uploading of images and system status information. To tackle the issues of small-object detection and limited edge-device computing power, an improved Dynamic-YOLOv8n model is proposed by introducing dynamic convolution and attention mechanisms to enhance recognition accuracy for small debris. Field experiments show that the system operates stably at vehicle speeds of 40–70 km/h, achieving an average detection accuracy of 93.2%. The results demonstrate that the proposed system achieves lightweight, real-time, and high-precision detection performance, providing an efficient and practical solution for road safety monitoring and digital maintenance.

Urban Digital Twins (UDT) require systematic integration of heterogeneous 3D geospatial data sources, but existing integration

methods struggle with semantic information loss during fusion, geometric precision degradation through format conversions, and

limited storage scalability. This paper presents a modular, database-centric framework achieving bidirectional semantic enrichment

through semantically enriched voxelization. The framework integrates CityGML building models, Mobile Mapping System (MMS)

point clouds, and Digital Terrain Models (DTM) using PostgreSQL/PostGIS database system with pgPointCloud, the point cloud

extension of Postgres for patch-based storage. A two-stage refinement pipeline is applied to align MMS point clouds to CityGML

wall surfaces using Random Sample Consensus (RANSAC) and Iterative Closest Point (ICP) algorithms. To integrate the terrain,

Constrained Delaunay Triangulation (CDT) algorithm is applied with building footprints as constraints. All datasets are independ-

ently voxelized at a common configurable resolution, with voxels enriched via custom pgPointCloud schemas storing multi-source

attributes. A unified voxel table merges layers using priority-based conflict resolution. The framework is evaluated in terms of com-

putational performance, registration precision, and storage efficiency, demonstrating feasibility and correctness of the integration

pipeline on a representative urban test case. This paper presents a proof-of-concept evaluated on a small urban area in Hannover,

Germany, demonstrating the framework’s potential for further development.

The increasing frequency and magnitude of landslides necessitates a fundamental shift from reactive mitigation to proactive, predictive risk governance. To define the necessary tools for this transition, this study conducts a systematic literature review and operational analysis of current Digital Twin (DT) implementations in the geosciences. Through this review, we identify four primary target user groups (emergency responders, technical experts, public administrators, and citizens) and map their specific 4D data requirements and interaction logics. Our findings highlight that most existing systems function as "Digital Shadows" characterised by unidirectional data flows and a topography gap, where dynamic sensor data is superimposed onto static, outdated 3D meshes. Based on these requirements, we propose a theoretical layered architectural framework for a Data Hub designed to bridge these gaps. The conceptual architecture is structured into three interconnected tiers: an Acquisition Layer for multi-scale data ingestion; a Modelling and Processing Layer for AI and physics-based stability assessment; and an Application and Service Layer for translating complex data into actionable intelligence. Finally, this work investigates a possible implementation path for landslides DT projects by outlining technical recommendations. This includes the adoption of cloud-native formats (e.g., Cloud Optimized GeoTIFF, Zarr) and unified interoperability standards (e.g., OGC SensorThings API) to evaluate the feasibility of transitioning towards a true bi-directional cyber-physical system for landslide risk management.

Highly Pathogenic Avian Influenza (HPAI), particularly the H5N1 strain, poses a significant ongoing threat to animal health, biodiversity and food security across Europe. Understanding where and when avian influenza risks intensify is essential for targeted surveillance and rapid response. This study develops a data-driven spatio-temporal framework that integrates geospatial, ecological and climatic datasets to explain and forecast the dynamics of H5N1 outbreaks between 2021 and 2024. Weekly country-level outbreak counts (208 weeks, 37 countries) were analysed using a hierarchical endemic-epidemic model with an assumption of Negative Binomial distribution. Environmental covariates, bird-species densities, and human population metrics were incorporated into endemic and autoregressive components. Model performance was evaluated using rolling one-step-ahead forecasts assessed by proper scoring rules (logarithmic score and ranked probability score) and calibration diagnostics. The proposed framework substantially outperformed a regression-only Negative Binomial baseline, reducing mean logS by approximately 29% and RPS by 49%, while exhibiting improved probabilistic calibration. Results indicate that H5N1 transmission is structured by ecological drivers and local persistence mechanisms rather than purely seasonal effects. Anseriformes, Charadriiformes and Pelecaniformes densities were identified as the key migratory bird families contributing to the viral spread. The endemic-epidemic model achieved high forecast accuracy, with majority of the of observed weekly outbreak counts falling within central predictive intervals (RPS = 0.76, logS = 0.61). Overall, the proposed framework provides a scalable approach for integrating ecological and spatial information into early-warning systems for HPAI surveillance.

Reliable communication between satellites and ground control stations (GCS) is fundamental to modern space missions, with its effectiveness being directly dependent on an unobstructed Line-of-Sight (LoS). Traditional site planning methods, relying on low-resolution terrain models, often overlook crucial obstacles like buildings or dense vegetation. This paper presents a comprehensive methodology using high-resolution Light Detection and Ranging (LiDAR) data, acquired from an Unmanned Aerial Vehicle (UAV), to precisely model horizon obstruction and optimize the placement of transceiver antennas. The methodology was verified on a real-world case study in Zielona Góra, Poland. The workflow included data acquisition, PPK-based trajectory processing, and point cloud subsampling using an Octree-based algorithm. The core of this work was the implementation of an algorithm to generate detailed elevation masks by calculating the maximum obstruction angle for defined azimuthal intervals. The analysis clearly identified the superior of two potential locations, proving the method's effectiveness as a decision-support tool in the space sector.

This paper investigates whether integrating the Microsoft Building Footprints (MBF) dataset with building footprints contributed by the OpenStreetMap (OSM) community can improve the spatial quality of building data. Specifically, the authors assess whether the resulting hybrid dataset enhances completeness and positional accuracy relative to the original MBF and OSM datasets. The evaluation was conducted in a study area encompassing both urban and rural environments, using 1:5,000 topographic cartography as the reference dataset. The merged MBF+OSM dataset successfully captured 87% of the buildings represented in the reference cartography, outperforming the standalone MBF and OSM datasets, which captured 81% and 70%, respectively. These results demonstrate that combining MBF and OSM footprints provides a more comprehensive representation of buildings and can offer a valuable alternative for applications requiring detailed, up-to-date building information.

The International Exposition of Turin held in 1911 in the Val-entino Park to celebrate the fiftieth anniversary of Italian unifi-cation; Hosted pavilions dedicated to science, industry, art, and architecture, symbolizing the modern spirit of post-unification Italy (Italy World’s Fairs, 2024). Today, only some traces of the project survive within the park. This disappearance has turned the exposition into a lost heritage landscape, known primarily through archival maps, photographs, and historical records. In late 2014 the project Turin 1911 started according to the cooperation between the Politecnico di Torino- Depart-ment of Architecture and Design in cooperation with the Uni-versity of California San Diego - School of arts and Humanities (https://italyworldsfairs.org/)

This project focuses on the digital revival of this exposition by integrating these materials with digital surveying and immer-sive visualization, aims to reproduce this vanished site and make it perceptible again to the public through virtual reality technologies.Within this framework, the research presented in this paper concentrates on the optimization of these mainly Revit and ArchiCAD modeled pavilions using the tools provid-ed by Unreal Engine for deployment on standalone VR sys-tems. The goal is to make heavy weight architectural scenes accessible in VR without connecting to PC and extending the concept to portable devices.

The integration of internet technologies and geographic information systems (GIS) has positioned Web Map Services(WMS) as indispensable tools for daily life, with their service quality attracting significant attention. This study proposes a quality evaluation model and method tailored for WMS, encompassing four critical dimensions: query and retrieval, map display, thematic services, and productization services. Empirical validation was conducted through functional, performance, and productization evaluations of three leading domestic platforms, utilizing technical benchmarks and user-centric metrics. Results demonstrate the model’s efficacy in quantifying service quality, aligning closely with real-world user experiences. The framework provides actionable guidelines for regulatory bodies to monitor service compliance and for providers to optimize architectural designs, thereby addressing gaps in personalization and cross-border functionalities observed in current systems. Furthermore, this work highlights the necessity of integrating emerging technologies—such as real-time traffic data and AI-driven personalization—to meet evolving demands for energy efficiency, global connectivity, and hyper-localized services. By bridging technical assessments with practical governance needs, the study offers a strategic roadmap for advancing service quality, supporting the development of China’s digital economy, and enhancing societal well-being through reliable geospatial solutions.

Virtual reality (VR) offers important advantages in training complex spatial skills, as required for example in driving, because it enables immersion and experience-based learning, and offers financial, ecological, and safety benefits. In the context of driving, as larger vehicles can be especially challenging to master for beginners, we investigated whether truck driving instruction and practice in a VR-simulator enhances performance, and whether the acquired skills transfer to maneuvering a real vehicle. In an empirical feasibility study, we first measured learners’ performance while an experienced instructor trained them on a conventional simulator vs. in VR, following analogous training protocols. The task was to reverse a truck with a trailer, a particularly difficult task that requires extensive practice. After training, participants completed a test on a real vehicle to validate the effectiveness of the training. Participants were asked to report previous experience, attitudes towards the system, motion sickness, and fatigue levels. Four male participants, who had a car driving license but no experience reversing a truck with trailer, completed the training. Results demonstrate that basic maneuvering skills can be trained in VR and transfer to the real vehicle. Even with a low-budget VR solution, participants learned easily, and learning curves were comparable to the simulator condition. Participants reported positive attitudes towards the training in both conditions. Future research could investigate whether using a customized VR environment that takes full advantage of all the benefits of VR, could lead to even greater training gains.

In recent years, the increasing frequency of natural disasters in remote and rugged areas has underscored the importance of unmanned aerial vehicles (UAVs) for rapid emergency response. This paper presents a novel approach for optimizing the placement of UAV docking stations in mountainous terrain for emergency operations. We develop a comprehensive, 3D Geodata framework that integrates 3D Digital Elevation Models (3D DEM), building infrastructure, and road network data to create a realistic three-dimensional optimization environment. The proposed system employs an Enhanced Adaptive Particle Swarm Optimization (EAPSO) algorithm with adaptive parameters, diversity maintenance mechanisms, and intelligent convergence detection to effectively handle the complex constraints of mountainous environments. Experimental results demonstrate that our 3D-aware EAPSO approach achieves superior performance in balancing coverage efficiency, energy consumption, and network connectivity compared to conventional optimization methods. The proposed system provides a scientific foundation for improving emergency response capabilities in challenging geographical environments.

The Sustainable development goals and the United Nations Ocean Decade require preservation of the oceans and the efficient management of marine resources, contributing to a sustainable oceans and blue economy. Oceans span a wide area with exclusive economic zones of different countries adjacent to each other. This therefore necessitates the collaborative management of these resources across several countries in a region. Data is essential to providing trusted information, which in turn drives knowledge generation from science to policy implementation, towards informed decision making regarding the ocean resources. Harmonising data into decision support tools becomes a challenge due to two main reasons. Firstly, due to the transdisciplinary nature of the ocean, where data is governed by a variety of standards. Secondly, regional collaboration requires the data and knowledge to be shared in a federated environment in order to preserve data sovereignty, while cognisant of the network challenges in developing countries. This paper presents a standard compliant framework for enabling data sharing and access in these environments based on lessons learnt in the Marine and Coastal Operations for Southern Africa and Western Indian Ocean region, a project supported by the African Union Commission‘s GMES and Africa program.

With the deep integration of big data and artificial intelligence technologies, the knowledge graph has emerged as an important method for organizing and understanding complex spatiotemporal information. Traditional knowledge graph management systems often face three significant challenges when dealing with spatiotemporal information in domains such as natural resource, urban studies, and emergency management. Firstly, the limited visualization capability makes it hard to intuitively represent the spatial distribution and temporal evolution of spatiotemporal knowledge. Secondly, the lack of systematic and deep machine-interpretable representation methods leads to inadequate diagnostic, predictive, and decision-making knowledge services. Thirdly, the knowledge construction process heavily relies on expert involvement resulting in high barriers to entry and low efficiency. To address these issues systematically, this paper designs and implements a comprehensive spatiotemporal knowledge graph construction and management system that integrates full lifecycle knowledge management, multi-form visualization methods, general and thematic knowledge graph construction.

Flooding is one of the most persistent natural hazards in Ghana, causing recurrent damage to infrastructure, livelihoods, and local economies. Despite its widespread impacts, most flood-related research has been concentrated on Accra, leaving many vulnerable regions understudied. This paper integrates Earth Observation (EO) datasets to identify and characterise flood-prone areas across Ghana at a national scale. Precipitation patterns between 2015 and 2025 were derived from the IMERG dataset, while Sentinel-1 Synthetic Aperture Radar (SAR) imagery was used for flood mapping through a change detection (ratio) approach. Results show a clear seasonal cycle, with major rainfall peaks from April to October, directly corresponding to observed flood events. Flooding is concentrated in the southern half of the country, particularly in Western, Western North and Eastern Regions, and recurrent hotspots around Kumasi in Ashanti and the Weija dam in Greater-Accra regions. Spatial patterns align closely with national topography, confirming the vulnerability of low-lying settlements and riverine communities. Technically, the study demonstrates the effectiveness of SAR-based change detection for flood mapping in data-sparse environments, while highlighting limitations related to in-situ validation and urban misclassification. From a policy perspective, the findings provide evidence to support flood risk management strategies, including targeted infrastructure investment and improved drainage planning. The results underline the necessity ofadopting engineering solutions to reduce flood vulnerability in communities in Ghana.

This contribution presents a feasibility study on representing access easement rights in multi-owned buildings using BIM and the IFC standard. A 3D BIM model was generated from 2D cadastral plans, and access easements between parking and storage units were modeled as explicit IFC entities with legal attributes such as beneficiary, servient unit, and restriction semantics. The study demonstrates how embedding easements as identifiable objects in IFC can enhance the clarity of Rights, Restrictions, and Responsibilities (RRRs) and improve the communication of legal constraints for future 3D digital cadaster applications.

The aim of the presented study is to develop and implement the strategy for digitally reconstructing and presenting the forgotten heritage associated with underground and open-pit mining conducted in the nowadays bilateral UNESCO Geopark Muskau Arch located on the border of Germany and Poland.

The research is led by scientific partners from Poland (Wroclaw University of Science and Technology) and Germany (Freiberg Technical University), with cooperation from the European Grouping of Territorial Cooperation (EGTC) Geopark Muskau Arch within the project “Digital Journey through Geopark Muskau Arch” co-financed from the European Regional Development Fund as part of the Poland-Saxony 2021-2027 INTERREG Cooperation Program.

Effective management of environmental safety risks in brownfield redevelopment relies on accurate identification of contaminated enterprises. A key challenge is the rapid acquisition of data on these enterprises. This study proposes a method leveraging Point of Interest (POI) data and a BERT-based prediction model to identify potentially contaminated enterprises. The method was applied to Datong, a major industrial and mining city in China. The method successfully identified 329 potentially contaminated enterprises across 23 types of polluting industries. Notably, enterprises in the mining and washing sectors of bituminous and anthracite coal represented 26.2% of the total identified, reflecting Datong’s coal-centric industrial nature. The proposed method efficiently identifies potentially contaminated enterprises, supporting targeted environmental risk management and brownfield redevelopment. Integrating it with regulatory frameworks can enhance compliance monitoring and inform decision-making for sustainable urban development.

The development of successful geospatial artificial intelligence (GeoAI) systems is hampered by two major obstacles: a scarcity of high-quality, annotated satellite imagery and a lack of unified platforms for modeling and testing. We introduce a scalable GeoAI framework that allows users to query, retrieve, and analyze high-resolution imagery using natural language interaction and direct processing of images. The system incorporates IBM-NASA's Prithvi Foundation Model for supervised detection of environmental features and the Clay Foundation Model for unsupervised similarity search when detectors are unavailable. An interactive interface allows users to search for features (such as swimming pools, vegetation changes, and burn scars), apply detectors to TIFF images, and explore new regions for model training

India, as one of the fastest-developing nations, faces severe air quality challenges due to rapid urbanization, industrialization, vehicular emissions, and agricultural activities. With major cities frequently exceeding WHO pollution limits, understanding the spatial and temporal behavior of atmospheric pollutants has become crucial. The integration of satellite-based geospatial technologies provides a powerful framework for assessing land–atmosphere interactions and their environmental implications. This study investigates the relationship between atmospheric pollutants and land surface characteristics across India using Sentinel-5P and Sentinel-2 datasets. The objective is to examine how pollutants influence vegetation health and urbanization through indices such as the Normalized Difference Vegetation Index (NDVI) and the Normalized Difference Built-up Index (NDBI). Google Earth Engine (GEE) and MATLAB were employed for data processing, statistical analysis, and visualization. NDVI and NDBI were derived from Sentinel-2 bands, while pollutant data (NO₂, SO₂, CO, O₃, HCHO, and CH₄) were extracted from Sentinel-5P. Correlation analysis, univariate regression, and temporal trend models were used to evaluate pollutant behavior and its linkages with land cover dynamics from 2019–2024. Results revealed strong positive correlations among NO₂, CO, SO₂, and HCHO (r = 0.59–0.76), indicating common anthropogenic sources, while NDVI showed significant negative correlations with O₃ (r = –0.46) and HCHO (r = –0.64). Formaldehyde and methane displayed the strongest increasing trends, highlighting growing emissions and vegetation response contrasts. The findings emphasize the interconnectedness of pollution, vegetation degradation, and urban expansion. Future research should integrate meteorological parameters and predictive modeling to strengthen sustainable environmental management and urban planning frameworks in India.

Based on the current situation of the application and service of national fundamental surveying and mapping achievements from 2020 to 2024, this paper adopts a combined method of quantitative and qualitative analysis to identify the existing problems and challenges, including constraints imposed by confidentiality management policies, the need to improve the timeliness and category diversity of data, and the insufficient service awareness and informatization service level. Corresponding countermeasures and suggestions for promoting the efficient provision and extensive utilization of fundamental surveying and mapping achievements are put forward, mainly including improving the policy and institutional system for the confidentiality management of surveying and mapping achievements, perfecting the achievement update mechanism, enriching the variety of achievements, advancing the processing and compilation of public-version surveying and mapping achievements, and constructing a public geographic information data innovation and application laboratory.

LoD-2 building models are more informative and practically more useful than LoD-1 representations because they capture the roof structure that defines the essential three-dimensional form of a building. They are important for applications such as urban planning, environmental simulation, and digital heritage. Although recent roof shape extraction methods can derive vectorised 2D roof structures from very-high-resolution imagery, transforming these image-based representations into fully textured 3D buildings remains challenging. In this paper, we present a semi-automated LoD-2 reconstruction pipeline that integrates HEAT-derived roof geometry with airborne LiDAR, satellite and Google Street View imagery. The 2D outputs are reprojected into map coordinates, fused with LiDAR through a two-stage roof reconstruction strategy to derive roof shapes and combined with an adaptive, LiDAR-based ground base initialisation to create a complete 3D wireframe. Roofs are textured using VHR orthophotos while the walls are textured via a process of Street View panorama selection, geometric filtering, Mask2Former segmentation, and homography rectification. Across a large-scale evaluation on 1000 buildings, the proposed two-stage reconstruction strategy improves geometric agreement with the LiDAR reference data achieving a roof-surface RMSE of 0.445~m. The wall texturing process produces convincing facades when suitable panoramas are available. While minor challenges such as sensitivities to LiDAR outliers, incomplete roof geometry, and facade occlusions persist, this pipeline effectively bridges 2D roof parsing and textured LoD-2 model generation, providing a robust and scalable foundation for advancing toward fully automated workflows.

This paper presents an automated framework for generating semantically labelled building point clouds from their corresponding BIM models. The proposed methodology aims to facilitate the creation of training datasets for deep learning–based indoor semantic segmentation. Two complementary labelling strategies are introduced. The first relies on bounding boxes (BBX) extracted from BIMelements to efficiently assign labels to points based on volumetric inclusion. The second approach uses solid geometry and a nearest-neighbour principle (SG-NN) to compute distances between BIM object meshes and the point cloud, enabling a more precise spatial correspondence. In addition, a room-based geometric grouping strategy is proposed to structure the annotated point clouds into spatial units compatible with common indoor segmentation datasets. The methods are evaluated through a qualitative analysis on several real building datasets of different typologies and acquisition conditions, as well as through a quantitative evaluation based on a manually segmented reference point cloud. Results show that the SG-NN approach achieves higher performance, with an average Recall of 92% and IoU of 88%, compared to 87% of Recall and %78 of IoU for the BBX approach. While the BBX approach provides faster processing, the SG-NN strategy achieves higher labelling accuracy, particularly for geometrically complex elements. The proposed workflow enables scalable dataset generation from Scan-to-BIM projects while significantly reducing manual annotation effort.

This paper presents an enhanced building extraction framework that combines deep learning-based image segmentation with photogrammetric point cloud refinement for urban roof detection. The method first applies a modified SegNet model to orthophotos from the ISPRS Vaihingen dataset to generate initial building masks. These results are then refined using geometric information from point clouds through ground filtering, clustering, and normal-guided region growing. By integrating spectral information from imagery with structural cues from 3D data, the proposed framework improves roof boundary delineation and reduces spurious detections. Experimental results on Areas 35 and 37 show that the method achieves strong overall performance, with a precision of 0.96, recall of 0.81, IoU of 0.78, and F1-score of 0.88. The findings indicate that point cloud refinement helps produce cleaner and more reliable building objects than image-based segmentation alone, especially in complex urban scenes. However, the approach remains sensitive to the density and quality of the point cloud. Overall, the study demonstrates that fusing orthophoto segmentation with point cloud processing is an effective strategy for more accurate and geometrically consistent building extraction.

The rapid expansion of urban areas and the continuous demand for their monitoring make remote sensing data a highly valuable tool for collecting large volumes of geospatial information in a relatively short time and with high repeatability. The main objective of this paper is to examine the potential offered by different types of geospatial data, as well as the relationships based on their scope, in comparison with measured reference data.

Architectural inventory tasks are useful not only for engineering projects but also for broader applications, such as environmental impact assessments, spatial planning, and related fields. This article introduces a rapid and cost-effective mixed-mode data collection framework for building inventory development, integrating terrestrial laser scanning, UAV imagery, and traditional ground measurements.

The paper will discuss the latest measurement technologies and their practical applications in building surveying, illustrated with a selected case study. The criteria for selecting appropriate measurement methods will also be analyzed, depending on the investor’s requirements and the intended use of the documentation.

This paper presents a set of techniques for updating the geometric information of buildings using laser scanning and imagery. It begins with an introduction to the fundamental concepts, terminology, and principles of 3D information. Subsequently, various measurement techniques are described, along with a discussion of potential sources of error and data incompleteness. The extracted geometric values are validated against independent survey data.

Accurate 3D reconstruction of urban scenes from multi-view images is essential for city planning, digital twins, and autonomous navigation. Traditional dense image matching relies on low-level cues such as intensity or gradients, which often produce noisy or incomplete point clouds in complex urban environments. This study introduces an attribute-enriched dense matching framework that embeds both geometric features and semantic attributes from multi-view images to guide dense image matching.

The framework first extracts semantic labels and geometric feature correspondences to generate intermediate products: conjugate features, feature seeds, an attribute map, and an initialized disparity map. These elements provide reliable priors that constrain dense matching, reduce search ranges, and prevent mismatches across structural boundaries. Dense image matching then propagates these constraints, producing an attribute-enriched disparity map and point cloud in which each 3D point carries both geometric and semantic information.

Evaluated on urban datasets, the proposed approach improves corner and edge localization, enhances edge continuity, reduces outliers in low-texture areas, and preserves semantic and structural attributes throughout 3D scene reconstruction. By integrating feature-based initialization with attribute-enriched dense image matching, the method delivers more accurate, interpretable, and robust 3D urban reconstructions, supporting downstream tasks such as precise measurement, object recognition, and scene analysis.

The rapid acquisition of high-precision parametric railway alignment is a fundamental prerequisite for intelligent railway construction and maintenance. Traditional measurement techniques and alignment fitting methods heavily rely on manual operations, often resulting in inefficiency, high costs, and insufficient accuracy control. To address these challenges, this study proposes an automated method for extracting and optimizing railway alignment from UAV-based LiDAR point clouds. Initially, track centerlines are extracted by leveraging the geometric smoothness of the railway and the structural characteristics of the track. A multi-constraint energy model integrating distance, orientation, and curvature is constructed to fit the geometric parameters of alignment elements, thereby providing high-quality initial values for subsequent alignment engineering parameter optimization. Finally, a global optimization strategy based on the simulated annealing algorithm is applied to jointly refine the engineering parameters of the standardized alignment composition, ensuring strict compliance with railway design specification. Experimental results demonstrate that the proposed method can efficiently and robustly extract high-precision alignment parameters with well-defined engineering semantics from complex railway point clouds, thereby providing reliable technical support for intelligent construction and full lifecycle management of railway systems.

Hyperspectral images (HSIs) are often severely degraded by mixed noise, such as Gaussian, stripe, and impulse noise during acquisition and transmission, which seriously impedes their subsequent applications. Therefore, HSI denoising is both crucial and challenging. In this work, we present a gradient-domain outlier-guided texture-preserved total variation (OG-TPTV) regularizer designed to remove mixed noise in HSIs. First, we utilize the mode-3 low-rank property of HSI gradient maps along the spectral dimension and apply a low-rank decomposition model to extract their spatial representation coefficients (SRCs). To improve the sparsity characterization of SRCs in the gradient subspace, an outlier-guided strategy is introduced. Specifically, we perform outlier detection on gradient maps to distinguish noise from texture structures and remove outliers to generate precise texture weighting maps. The resulting texture weight maps offer adaptive guidance for adjusting the strength of the sparsity constraints. Finally, a denoising method for HSIs is developed based on OG-TPTV. Extensive experiments on both synthetic and real HSIs demonstrate the superior denoising performance of our method.

Hyperspectral satellite missions generate massive data volumes that are difficult to transmit and store under tight onboard resource constraints, making effective lossy compression a key enabling technology. We propose SpectralNet-X, a transformer-based autoencoder for spectral-only compression of spaceborne hyperspectral imagery at a fixed compression ratio of 16. The encoder maps each spectrum to a low-dimensional latent code using a 1D convolutional projection followed by stacked self-attention layers with rotary position embeddings, and aggregates information via cross-attention pooling. The decoder reconstructs full-band spectra through an upsampling stack and per-band affine calibration. To improve reconstruction fidelity and generalization, SpectralNet-X is first pretrained in a masked-signal reconstruction task inspired by SimMIM and then fine-tuned with a mixed objective that combines mean-squared error and spectral angle mapper (SAM) terms using a scheduled weighting scheme. We evaluate SpectralNet-X on the large-scale HySpecNet–11k benchmark and in a mission-realistic cross-sensor setting, where models trained on HySpecNet–11k are tested on PRISMA hyperspectral scenes. Across PSNR, SSIM, and SAM, and when compared to three different compression autoencoders, SpectralNet-X achieves the lowest angular reconstruction errors while maintaining competitive distortion metrics and substantially reducing the fraction of spectra with large SAM outliers. These results indicate that transformer-based spectral compression is a promising candidate for robust, mission-realistic onboard hyperspectral data reduction.

Validating the performance of deep learning models in satellite imagery is essential for ensuring model generalizability, decision reliability, and spatial transferability—particularly in the context of hyperspectral images, which contain high-dimensional, spatially complex data. While it is well recognized that multiple spatial characteristics influence deep learning model performance, few studies have systematically examined how the interactions among these characteristics affect model validation sensitivity in hyperspectral contexts. This study aims to investigate how the interaction between spatial scale (e.g., surrounding 3, 5, 7 grids) and training sample size (e.g., 10%, 30%, 50% of all data) influences the validation accuracy and sensitivity of deep learning models. An innovative validation sensitivity index is developed to quantify the change in accuracy per unit of spatial scale and sample size, enabling a more refined assessment of model robustness. The index is applied to three representative hyperspectral datasets, covering diverse environmental and spectral conditions. Results show that spatial scale accounts for 0~21.0% accuracy variation, training sample size contributes 5.6~36.5% variation, but their interaction leads to 5.4~70.3% variation, indicating a nonlinear amplification enhanced effect. These findings may be explained by the compounded influence of data contextuality, spatial redundancy, and model overfitting dynamics. This study demonstrates the critical need to consider spatial interactions in validation design, offering new insights for enhancing the reliability of geospatial artificial intelligence (GeoAI) applications in remote sensing and spatial data science.

Surface reflectance is a foundational variable in optical remote sensing, as inaccuracies introduced during atmospheric correction can propagate and amplify across subsequent satellite-derived products. Nonetheless, the extent to which the choice of Radiative Transfer Model (RTM) affects reflectance retrieval has not been sufficiently examined. This study investigates how two widely used RTMs—6SV and VLIDORT—produce different surface reflectance outcomes when applied under consistent atmospheric and geometric conditions for the GEO-KOMPSAT-2B/GEMS instrument. To ensure comparability, both models were driven by identical GEMS aerosol properties and an equivalent LUT configuration.

The comparison shows that while the two RTMs reproduce broadly similar spatial patterns, systematic quantitative differences remain in the retrieved reflectance. These differences vary depending on atmospheric and viewing conditions, particularly under higher aerosol loading. A sensitivity analysis further indicates that aerosol amount and scattering characteristics, alongside viewing geometry, are key factors influencing the magnitude of RTM divergence.

Overall, this study provides a structured assessment of RTM-dependent variability in atmospheric correction and highlights the importance of model choice when interpreting or harmonizing surface reflectance products. The findings offer a basis for improving consistency in future GEMS-based retrievals and for advancing reliable surface reflectance generation in geostationary remote sensing.

Remote sensing acquisition systems rely on a range of platforms, from drones to satellite missions, to record multimodal Earth surface data. This fact encourages the preparation of datasets with complementary properties, thereby increasing their discriminative potential. A common complementary combination is between Hyperspectral and LiDAR-generated digital surface model data. While engaging, this fusion poses challenges for specific applications. Multiple works fuse these modalities at the feature level using vector concatenation, maximization, or averaging. Although functional, these methods omit target interactions between the modalities. Another challenge in remote sensing is the quantity and quality of labels required by deep learning methods, which are expensive, error-prone, and difficult to scale. We address the challenges above by proposing a self-supervised processing framework based on cross-modal attention that effectively fuses features at multiple levels, thereby exploiting complementary information across data streams. Specifically, our method is founded on a pseudo-Siamese network that reweights each modality’s features with information from the other via a mirrored cross-modal attention. The network’s objective is to maximize the similarity between the feature representations of both streams. A fusion network builds a latent representation using the learned encoders and attention modules. Then, a k-Nearest Neighbor classifier categorizes each sample within the representation using ten labels per class. Our experiments show that our spatial- and channel-spatial cross-modal attention approaches outperform well-established fusion methods for label-efficient land cover classification across datasets. Our findings lay the groundwork for fusion methods that effectively exploit inter-stream data relationships to encourage complementarity.

Despite the rapid development of satellite sensors, acquiring high-resolution RGB images remains a challenge. In this paper, a GAN-based multiscale feature-based pan-to-rgb model is proposed to establish a novel framework for high-resolution, high-fidelity RGB images generation from remote sensing panchromatic images. The spatial structure, texture, and color of the results are consistent with the real images, and the colors are naturally realistic and vibrant. Multiscale features and symmetric luminance color decoders are utilized to overcome color desaturation, inaccuracy, and distortion in conventional algorithms. By combining CNNs for local feature modeling and transformers for global feature modeling, this approach learns pan-to-rgb mappings to produce high-resolution, high-fidelity RGB images in CIELAB space. Besides, the luminance distance loss and the color distance loss are utilized to prevent the coupling of luminance and color. We also conducted experimental validation on Gaofen-7 satellite data, and the results demonstrated that the FID, CF, and △CF indicators of the proposed algorithm improved by 2.90%, 11.77%, and 64.51%, respectively, compared to the comparison algorithms.

Miniaturized UAV laser scanning systems have advanced rapidly over the past decade, especially in the low-cost sector. DJI entered this field with the Zenmuse L-series, integrating GNSS/INS with compact scanners. While the first-generation L1 showed moderate precision, the L2 improved notably through reduced beam divergence. In November 2025, DJI released the Zenmuse L3. In this contribution, we assess its performance.

The main upgrade from L2 to L3 lies in the LiDAR unit: L3 uses a single 1535 nm laser instead of multiple 905 nm diodes, offers a symmetric 0.25 mrad beam divergence, and supports pulse repetition rates from 350 kHz to 2 MHz. High PRR operation is limited to altitudes ≤50 m due to missing multiple-time-around resolution. Scan modes include linear, non-repetitive, and a new star-shaped pattern.

L2 and L3 were tested at three sites in Lower Austria covering a warehouse, power-lines, and forests. Flights were conducted at 80 m AGL (350 kHz) and, for the warehouse, 50 m AGL (2 MHz). Precision, strip consistency, point density, feature separability, and vegetation penetration were evaluated using the scientific software OPALS.

L3 data showed sharper edges, reduced noise, and higher separability, yielding spline-fit residuals of 0.9 cm versus 2.6 cm for L2 for reconstructing a double-threaded power-line. Ground point coverage in forests increased from 18 % (L2) to 51 % (L3). Strip height differences are around 2 cm for both sensors and L3 achieved sub-centimeter precision on sealed surfaces. Overall, L3 offers substantial gains in spatial resolution, precision, and vegetation penetration.

Intrinsic calibration of rotating multi-beam lidars (RMBL) enables more precise measurements. We calibrated our sensor to improve its geometric and radiometric accuracy using a rotating tilted platform. The rotating mechanism widens the field of view of each lidar channel and allows all lasers of the sensor to measure the same areas in a room containing planar wall and floor sections. Therefore, we can collect measurements for geometric and radiometric calibration with minimal amount of calibration targets. Furthermore, we used data based numerical minimization to estimate the calibration parameters for all 128 lidar channels in our RMBL sensor. For the intrinsic geometric calibration of the sensor, we estimated the elevation and azimuth angles of each laser. For the radiometry, we estimated a linear model for each laser to correct the intensity measurement. For a linear model, two different known diffuse reflectance targets are sufficient for the radiometric calibration. We tested our methods in two different environments, in an office room and a longer corridor. We showed that the methods can improve the precision of the RMBL sensor significantly. Regarding geometry, we were able to reduce the error on average from 16.1 mm to 15.1 mm (6.2% improvement). For radiometry, we were able to improve the reflectance measuring accuracy on average from 9.5% errors down to -0.9% errors (91% improvement).

In recent years, laser scanning has evolved into a core surveying technology for 3D mapping, both statically from stationary scan positions (terrestrial laser scanning, TLS) and kinematically from moving platforms (kinematic laser scanning, KLS). Consequently, there is a growing demand for methods that efficiently and coherently support both static and kinematic data acquisition modes. This contribution presents a tightly-coupled approach for the co-registration of TLS and KLS data, which simultaneously integrates GNSS positions, inertial measurements, planar features extracted from both static and kinematic point clouds, and control information in a joint non-linear least-squares adjustment. This is neither just a transformation of the kinematic onto the static point cloud nor a simple correction of the trajectory in e.g., a strip adjustment, but rather a tightly coupled adjustment of static and kinematic data. This approach avoids the need for additional survey control for kinematic data by leveraging the static scan data as a proxy, enabling accurate georeferencing even in scenarios where the individual datasets cannot be reliably tied to control points. Results show that the co-registration notably improves the relative consistency of kinematic datasets with respect to a static reference. Such co-registration enables new use-cases for multi-modal data acquisition, such as change-detection in repeated kinematic data acquisitions with respect to a static reference dataset, or more flexible ways of integrating ground control in kinematic surveys.

This study investigates the use of a distributed asynchronous lidar system for augmented position and orientation determination in Global Navigation Satellite Systems (GNSS) denied environments. An asynchronous lidar design is one in which the laser transmitter and detectors/receivers are disconnected and carried on separate platforms. This unique geometry offers observational redundancy that can be used to estimate the trajectory of the receiver platforms. The paper presents the results of simulation experiments, first examining single epoch solutions and then considers estimates of position and orientation along simulated flight trajectories. The results show that as long as the laser transmitter is operated above the GNSS denied environment, the system is able to simultaneously estimate position and orientation for multiple receiver drones, even for extended periods of GNSS outages. The accuracy of position and orientation estimation is dependent on the exact flight path and the number of lidar receivers in the solution, but with favorable geometry the accuracy of position estimation can approach that provided by a high precision GNSS solution.

This work extends our initial proof-of-concept via emulations on the benefits of relative spatial constraints between imagery and lidar point clouds in a factor graph based optimization with satellite positioning (GNSS) and raw inertial readings (Mouzakidou et al., 2025). Here, we demonstrate practically the automatic extraction and integration of 2D-3D correspondences established in the 3D domain within rough natural terrain flown over by an aircraft with sensors of high quality. We show that considering cross-domain (i.e. 2D-3D) constraints enables the calibration of internal camera parameters and its boresight on job, i.e. within mapping flight configurations, where conventional approaches fail. The common optimization of raw IMU data with such constraints improves the respective agreements between the lidar and image dense clouds, achieving consistency at ground resolution level, which is not the case for the conventional (standard) processing of acquired data.

Visual-inertial odometry (VIO) plays a key role in modern navigation and mapping systems.

For their successful integration, an initialization phase, in which IMU-related bias factors are estimated, becomes a fundamental step.

Without one, the subsequent nonlinear estimation of the platform pose may fail to converge or completely diverge.

As reliance on visual and inertial information may exhibit instability due to error accumulation with time, incorporating absolute positioning information through global navigation satellite system (GNSS) measurements, may enhance its robustness and accuracy.

Accordingly, GNSS and visual-inertial initialization frameworks have been receiving growing attention in recent years where current strategies tend to follow a loosely-coupled formulation that first initializes the VIO trajectory, and then aligns it with GNSS measurements.

Such strategies are multi-stage, nonlinear, and computationally expensive, motivating us to introduce an alternative framework in which GNSS position is integrated with the raw visual-inertial measurements to form absolute translation constraints.

Accordingly, we achieve a closed-form, linear and globally consistent drift-free solution which is computationally efficient and requires neither 3D reconstruction nor nonlinear refinement, as common approaches do.

Testing our initialization formulation on benchmark multi-sensor datasets, results show that we outperform current baselines while exhibiting robustness in challenging scenarios.

Artificial Intelligence (AI) integration has become an essential of modern AEC workflows, yet it has failed to gain a position in waste management. This gap is particularly prominent given the urgent environmental and legal imperatives for the sector to mitigate its demolition outputs. Existing approaches to waste classification and diversion cost estimation rely on manual interpretation of project documentation, a process that is both resource-intensive and structurally incompatible with the machine-readable data environments established by Building Information Modelling (BIM). This paper presents a framework that bridges Industry Foundation Class (IFC) compliant BIM data and Large Language Model (LLM) capabilities to automate Construction and Demolition Waste (C&DW) classification and probabilistic cost optimisation. The framework utilizes IfcOpenShell to extract element geometry and material data, channeling this information into a Retrieval-Augmented Generation (RAG) pipeline. To ensure rigorous compliance during classification, a FAISS-indexed knowledge base grounds a locally deployed Llama3 model against the specific mandates of Province of Ontario, Canada regulation 102/94. Diversion cost scenarios are computed through a Bayesian cost module coupled to a multi-objective genetic algorithm (MOGA) optimiser. Th proposed approach is evaluated against a labelled dataset of 104 IFC type-and-material combinations, the RAG classifier. Performance thresholds were established a piori based on multi-class classification benchmarks and Bayesian cost model uncertainty tolerances. The framework achieved a macro-average F1 of 0.84 and overall accuracy of 88%, satisfying the minimum criteria for automated C&DW characterization under Ontario Regulation 102/94.

This case study details the integration of official large-scale open 2D and 3D geospatial data of the city of Berlin, Germany, into the Virtual Battlespace 4 (VBS4) simulator for security applications. Realistic scenery with elements specific to the target area is obtained from a digital terrain model, true-ortho mosaic, and high-resolution land use/land cover layer rasterized from OpenStreetMap vector primitives. For the central Mitte borough with its government institutions and foreign embassies, almost 20000 buildings are prepared from textured CityGML data in an automatic multi-stage process. This process involves pre-wrapping the texture images, which are referenced by the semantic 3D models using non-canonical coordinates, and the rapid creation of compact atlases to reduce the bitmap count by three orders of magnitude. To ensure that the building meshes blend seamlessly into the terrain, vertical adjustment methods are discussed, and ground extrusion is implemented to approach the model's base surfaces from below. Data import into VBS4 happens through its Geo interface for the terrain, ortho, and land cover, while the buildings are compiled into an add-on with a custom workflow that involves reprojection, collision component setup, and damage behavior configuration. During interactive convoy training in the virtual environment, a high recognition value compared to the real landscape could be attested visually. Simulation exhibited acceptable frame rates, but required considerable computing resources.