High-resolution Imagery Research Articles

Beyond the Greater Angkor Region: Automatic large-scale mapping of Angkorian-period reservoirs in satellite imagery using deep learning

Archaeologists often use high-resolution satellite imagery to identify potential archaeological sites or features, including ancient settlements, burial mounds, roads, and even subtle differences in vegetation or topography. Over the last several decades, satellite imagery and other remote sensing techniques (including aerial photography and LiDAR data) have been used to thoroughly map the extensive settlement complex of the Greater Angkor Region (1 500 km2, 9th – 14th centuries CE) in present-day Cambodia. While we now have a comprehensive map of this area, the landscapes beyond the Greater Angkor Region that formed the Angkorian cultural sphere have not been mapped, even though the density of features on the landscape seems to continue beyond the area considered Greater Angkor. While a comprehensive settlement study of the entire Angkorian realm would be incredibly helpful in understanding patterns of ancient urbanism and early statehood in Southeast Asia, mapping this area using manual identification of archaeological features in satellite imagery would be highly time-consuming. In this paper, we employ a state-of-the-art deep learning model for semantic segmentation using Deeplab V3 + to identify one typical and characteristic feature: Angkor-period reservoirs. Our results indicate that this AI model is accurate enough to provide a valuable “second opinion” to landscape archaeologists to enhance and quicken their mapping process, making them substantially more productive. The deep learning model for semantic segmentation employed here, which can be trained on other types of archaeological and non-archaeological features worldwide, will be a valuable tool for areas of research that involve intensive manual investigation and interpretation of satellite imagery and will aid researchers as they continue to map the Angkorian world.

Remote Sensing of Forest Gap Dynamics in the Białowieża Forest: Comparison of Multitemporal Airborne Laser Scanning and High-Resolution Aerial Imagery Point Clouds

Remote sensing technologies like airborne laser scanning (ALS) and digital aerial photogrammetry (DAP) have emerged as efficient tools for detecting and analysing canopy gaps (CGs). Comparing these technologies is essential to determine their functionality and applicability in various environments. Thus, this study aimed to assess CG dynamics in the temperate European Białowieża Forest between 2015 and 2022 by comparing ALS data and image-derived point clouds (IPC) from DAP, to evaluate their respective capabilities in describing and analysing forest CG dynamics. Our results demonstrated that ALS-based point clouds provided more detailed and precise spatial information about both the vertical and horizontal structure of forest CGs compared to IPC. ALS detected 27,754 (54%) new CGs between 2015 and 2022, while IPC identified 23,502 (75%) new CGs. Both the average gap area and the total gap area significantly increased over time in both methods. ALS data not only identified a greater number of CGs, particularly smaller ones (below 500 m2), but also produced a more precise representation of CG shape and structure. In conclusion, precise, multi-temporal remote sensing data on the distribution and size of canopy gaps enable effective monitoring of structural changes and disturbances in forest stands, which in turn supports more efficient forest management, e.g., planning of forest regeneration.

Toward Reliable Post-Disaster Assessment: Advancing Building Damage Detection Using You Only Look Once Convolutional Neural Network and Satellite Imagery

Natural disasters continuously threaten populations worldwide, with hydrometeorological events standing out due to their unpredictability, rapid onset, and significant destructive capacity. However, developing countries often face severe budgetary constraints and rely heavily on international support, limiting their ability to implement optimal disaster response strategies. This study addresses these challenges by developing and implementing YOLOv8-based deep learning models trained on high-resolution satellite imagery from the Maxar GeoEye-1 satellite. Unlike prior studies, we introduce a manually labeled dataset, consisting of 1400 undamaged and 1200 damaged buildings, derived from pre- and post-Hurricane Maria imagery. This dataset has been publicly released, providing a benchmark for future disaster assessment research. Additionally, we conduct a systematic evaluation of optimization strategies, comparing SGD with momentum, RMSProp, Adam, AdaMax, NAdam, and AdamW. Our results demonstrate that SGD with momentum outperforms Adam-based optimizers in training stability, convergence speed, and reliability across higher confidence thresholds, leading to more robust and consistent disaster damage predictions. To enhance usability, we propose deploying the trained model via a REST API, enabling real-time damage assessment with minimal computational resources, making it a low-cost, scalable tool for government agencies and humanitarian organizations. These findings contribute to machine learning-based disaster response, offering an efficient, cost-effective framework for large-scale damage assessment and reinforcing the importance of model selection, hyperparameter tuning, and optimization functions in critical real-world applications.

Automated Detection of Pedestrian and Bicycle Lanes from High-Resolution Aerial Images by Integrating Image Processing and Artificial Intelligence (AI) Techniques

The rapid advancement of computer vision technology is transforming how transportation agencies collect roadway characteristics inventory (RCI) data, yielding substantial savings in resources and time. Traditionally, capturing roadway data through image processing was seen as both difficult and error-prone. However, considering the recent improvements in computational power and image recognition techniques, there are now reliable methods to identify and map various roadway elements from multiple imagery sources. Notably, comprehensive geospatial data for pedestrian and bicycle lanes are still lacking across many state and local roadways, including those in the State of Florida, despite the essential role this information plays in optimizing traffic efficiency and reducing crashes. Developing fast, efficient methods to gather this data are essential for transportation agencies as they also support objectives like identifying outdated or obscured markings, analyzing pedestrian and bicycle lane placements relative to crosswalks, turning lanes, and school zones, and assessing crash patterns in the associated areas. This study introduces an innovative approach using deep neural network models in image processing and computer vision to detect and extract pedestrian and bicycle lane features from very high-resolution aerial imagery, with a focus on public roadways in Florida. Using YOLOv5 and MTRE-based deep learning models, this study extracts and segments bicycle and pedestrian features from high-resolution aerial images, creating a geospatial inventory of these roadway features. Detected features were post-processed and compared with ground truth data to evaluate performance. When tested against ground truth data from Leon County, Florida, the models demonstrated accuracy rates of 73% for pedestrian lanes and 89% for bicycle lanes. This initiative is vital for transportation agencies, enhancing infrastructure management by enabling timely identification of aging or obscured lane markings, which are crucial for maintaining safe transportation networks.

In-Season Estimation of Japanese Squash Using High-Spatial-Resolution Time-Series Satellite Imagery

Yield maps and in-season forecasts help optimize agricultural practices. The traditional approaches to predicting yield during the growing season often rely on ground-based observations, which are time-consuming and labor-intensive. Remote sensing offers a promising alternative by providing frequent and spatially extensive information on crop development. In this study, we evaluated the feasibility of high-resolution satellite imagery for the early yield prediction of an under-investigated crop, Japanese squash (Cucurbita maxima), in a small farm in Hollister, California, over the growing seasons of 2022 and 2023 using vegetation indices, including the Normalized Difference Vegetation Index (NDVI) and the Soil-Adjusted Vegetation Index (SAVI). We identified the optimal time for yield prediction and compared the performances across satellite platforms (Sentinel-2: 10 m; PlanetScope: 3 m; SkySat: 0.5 m). Pearson’s correlation coefficient (r) was employed to determine the dependencies between the yield and vegetation indices measured at various stages throughout the squash growing season. The results showed that SkySat-derived vegetation indices outperformed those of Sentinel-2 and PlanetScope in explaining the squash yields (R2 = 0.75–0.76; RMSE = 0.8–1.9 tons/ha). Remote sensing showed very strong correlations with yield as early as 29 days after planting in 2022 and 37 and 76 days in 2023 for the NDVI and the SAVI, respectively. These early dates corresponded with the vegetative stages when the crop canopy became denser before fruit development. These findings highlight the utility of high-resolution imagery for in-season yield estimation and within-field variability detection. Detecting yield variability early enables timely management interventions to optimize crop productivity and resource efficiency, a critical advantage for small-scale farms, where marginal yield changes impact economic outcomes.

Preliminary investigation of hummocky landforms and hyper-mobility of the Bingda landslide, northeastern Tibetan Plateau

Rapid and long-runout landslides characterized by their high speed, long distance mobility, and huge capacity and volume would pose significant threats to infrastructure and life safety. In this study, a rapid and long-runout landslide that occurred in the Bingda village of the northeastern Tibetan Plateau, which was triggered by heavy rainfall in June 2017, was preliminarily investigated. On the basis of detailed field surveys, high-resolution satellite imagery analysis, and laboratory tests, the morphological and sedimentological features of the landslide were described, and the formation mechanism of hummocky landforms and its insight into the extraordinary movement of the Bingda landslide was deduced. The field investigation and satellite imagery analysis showed that there were nearly 200 hummocks, mostly with normal circular bases and with a height of ∼0.1 m–7.5 m, distributed in the transfer and accumulation areas of the landslide. The height and number density of the hummocks decreased away from the transfer area to the accumulation area and displayed higher heights at the outer bends of the gully channel than that at the inner bends of it. The characteristics of the spatial distribution and the composition of hummocks indicated that significant generation and dissipation of pore-water pressure within the loose and saturated silty clay layer in the runout path was the most probable reason for the formation of hummocky landforms. This study also provided insights into the hypermobility mechanisms of the Bingda landslide, suggesting that this landslide began with the sliding failure of the weathered colluvium in the source area, and then the landslide debris traveled into the channel and impacted sudden undrained loading and rapid shearing to the underlying silty clay layers in the gully. These processes generated pore-water pressure and reduced the effective stress within the soil particles, resulting in a decrease in the frictional resistance in the substrate, finally facilitating the rapid and long-runout movement of the landslide.

The Strategies to AEC Implement BIM-GIS-Drone-Mobile Apps-AI Towards Future Demands, Case Study: Setiu, Terengganu

In the Architecture, Engineering, and Construction (AEC) industry, the integration of Building Information Modelling (BIM), Geographic Information System (GIS), and Drone Mobile Apps has revolutionized how professionals plan, design, construct, and manage buildings and infrastructure. This integration offers a comprehensive approach that leverages the strengths of each technology to enhance efficiency, accuracy, and collaboration throughout the project lifecycle. BIM is a process that involves creating digital representations of physical and functional properties of buildings or infrastructure. GIS is a system that collects, analyses, and presents spatial data related to natural and built environmental assets such as wetland resources are being overused because of severe occurrences and conditions such the rapidly rising population, the rapidly improving economic climate, and the expanding metropolitan region. Drone Mobile Apps have become an integral part of modern construction projects due to their ability to capture high-resolution imagery efficiently using drones. Artificial Intelligence (AI) plays a crucial role in processing vast amounts of data generated by BIM models, GIS databases, drones, and mobile apps due to social trends. However, the systematic implementation of current advanced technologies in AEC projects in the wetland area experience challenges. This study aims to identify critical strategies to be implemented in the current digital transformation. The significance of this study is to support current trends of technology in RMK-12, 4.0 Digital Transformation. Initially, 20 strategies are drawn from the existing literature review and 10 critical strategies are identified focusing on the strategies for BIM and GIS Integration with AI interoperability. An in-depth interview with 33 experts as qualitative study and quantitative study using a questionnaire survey with 100 respondents from construction field. The results may improve project efficiency and accuracy while ensuring better decision-making processes throughout the project lifecycle of the AEC industry in the wetland area.

Integrative deep learning framework for analysing population density shifts and flood inundation patterns

ABSTRACT This study introduces a robust deep learning framework to analyze the relationship between population density changes and flood inundation, which is crucial for improving disaster management and urban resilience strategies. As the environment changes, population grows rapidly, and infrastructure advances, it is vital to ensure urban areas can withstand floods while maintaining functionality and safety. Preparedness through anticipation and resource allocation in high-risk areas enhances risk assessment. Using high-resolution satellite imagery, socio-economic datasets, and a modified U-Net architecture to process multispectral and Synthetic Aperture Radar (SAR) data, the study generates detailed maps of population shifts and flood extents. The Intersection over Union (IoU) metric rigorously validates the model's accuracy in predicting and mapping flood and population data. The findings show significant correlations between flood events and population distribution changes, offering empirical insights into infrastructure development and resilience planning. These insights help develop effective policies to manage urbanization and population in flood-prone areas. Additionally, the results provide data-driven insights for infrastructure and resilience planning in response to climate change challenges. This information aids policymakers in addressing the impacts of climate change on vulnerable communities, supporting community empowerment, and facilitating early restoration post crisis.

Adversarial Positive-Unlabeled Learning-Based Invasive Plant Detection in Alpine Wetland Using Jilin-1 and Sentinel-2 Imageries

Invasive plants (IPs) pose a significant threat to local ecosystems. Recent advances in remote sensing (RS) and deep learning (DL) significantly improved the accuracy of IP detection. However, mainstream DL methods often require large, high-quality labeled data, leading to resource inefficiencies. In this study, a deep learning framework called adversarial positive-unlabeled learning (APUL) was proposed to achieve high-precision IP detection using a limited number of target plant samples. APUL employs a dual-branch discriminator to constrain the class prior-free classifier, effectively harnessing information from positive-unlabeled data through the adversarial process and enhancing the accuracy of IP detection. The framework was tested on very high-resolution Jilin-1 and Sentinel-2 imagery of Bayinbuluke grasslands in Xinjiang, where the invasion of Pedicularis kansuensis has caused serious ecological and livestock damage. Results indicate that the adversarial structure can significantly improve the performance of positive-unlabeled learning (PUL) methods, and the class prior-free approach outperforms traditional PUL methods in IP detection. APUL achieved an overall accuracy of 92.2% and an F1-score of 0.80, revealing that Pedicularis kansuensis has invaded 4.43% of the local plant population in the Bayinbuluke grasslands, underscoring the urgent need for timely control measures.

Detection of Structural Damage After an Earthquake Using GIS and Remote Sensing Methods

Developments in Geographic Information Systems and Remote Sensing (RS) technologies and innovative approaches emerging in deep learning (DL) supported analysis methods have an important place in disaster research as in every field. Convolutional neural networks such as Mask RCNN, U-NET, one of the deep learning methods for disaster damage impact assessment and classification, have started to show successful results. However, high-resolution geospatial imagery and drones provide faster and more accurate detection of structural damage. In this study, damaged building detection was performed using Göktürk-1 satellite images from 6 February 2023 using Mask-RCNN architecture. In this study, deep learning methods were used to detect the collapsed buildings in the city of Malatya during the 6 February 2023 earthquakes. The study aims to emphasize the significance of GIS and remote sensing for the timely and efficient evaluation of building damage after a disaster. Considering this, high quality images of Malatya city before and after the earthquake were analyzed and data sets were created by masking using Mask RCNN deep learning method through ArcGIS Pro 3.4.0 software. According to the results of the research, it quickly detected damaged buildings with an accuracy rate of 70% according to satellite images after the earthquake. As a result, GIS and deep learning models were used to detect and map the initial damage after the earthquake.

Analysis of the impact of multiple green space patterns and key meteorological factors on particulate matter pollution: a case study in the Zhengzhou metropolitan area.

Atmospheric particulate matter (PM) is a primary pollutant affecting urban air quality, posing increasing threats to public health and ecological environments. While urban green spaces and meteorological conditions individually influence PM pollution, the mechanisms by which meteorological indicators mediate the relationship between green space patterns and PM concentrations remain unclear. We used daily PM concentration data in the Zhengzhou Metropolitan Area (ZMA) in 2021, combined with high-resolution satellite imagery and climate monitoring data. By employing Generalized Linear Models (GLMs) and Partial Least Squares Structural Equation Modeling (PLS-SEM), we investigated the effects of green spaces and meteorological conditions on PM, highlighting the significant mediating role of key meteorological indicators in the process by which green spaces mitigate PM pollution. Results indicated that PM2.5 concentrations were more sensitive to green space patterns and meteorological conditions at 1-6km scales compared to PM10. Significant scale-dependent differences were observed in the coupling between PM concentrations and green spaces. PLS-SEM revealed that key meteorological indicators, particularly wind speed and humidity, significantly mediated the impact of green spaces on PM pollution, with mediation effects peaking at the 4km scale. The percentage of largest green space patches had the most pronounced mediated effect on PM2.5 and PM10 through climate factors. Conclusively, to maximize ecological benefits, it is essential to consider wind speed and humidity around green spaces. The findings emphasize the importance of optimizing green space patterns at multiple scales and incorporating local microclimate considerations in future PM pollution management within the ZMA.

Coastline and Riverbed Change Detection in the Broader Area of the City of Patras Using Very High-Resolution Multi-Temporal Imagery

Accurate and robust information on land cover changes in urban and coastal areas is essential for effective urban land management, ecosystem monitoring, and urban planning. This paper details the methodology and results of a pixel-level classification and change detection analysis, leveraging 1945 Royal Air Force (RAF) aerial imagery and 2011 Very High-Resolution (VHR) multispectral WorldView-2 satellite imagery from the broader area of Patras, Greece. Our attention is mainly focused on the changes in the coastline from the city of Patras to the northeast direction and the two major rivers, Charadros and Selemnos. The methodology involves preprocessing steps such as registration, denoising, and resolution adjustments to ensure computational feasibility for both coastal and riverbed change detection procedures while maintaining critical spatial features. For change detection at coastal areas over time, the Normalized Difference Water Index (NDWI) was applied to the new imagery to mask out the sea from the coastline and manually archive imagery from 1945. To determine the differences in the coastline between 1945 and 2011, we perform image differencing by subtracting the 1945 image from the 2011 image. This highlights the areas where changes have occurred over time. To conduct riverbed change detection, feature extraction using the Gray-Level Co-occurrence Matrix (GLCM) was applied to capture spatial characteristics. A Support Vector Machine (SVM) classification model was trained to distinguish river pixels from non-river pixels, enabling the identification of changes in riverbeds and achieving 92.6% and 92.5% accuracy for new and old imagery, respectively. Post-classification processing included classification maps to enhance the visualization of the detected changes. This approach highlights the potential of combining historical and modern imagery with supervised machine learning methods to effectively assess coastal erosion and riverbed alterations.

Utilization of satellite imagery and artificial intelligence for disaster management: Approaches and case studies

The advent of advanced satellite observations and the rapid evolution of Artificial Intelligence (AI) technologies have led to a fundamental shift in disaster management. These technologies enhance precise prediction, closer monitoring, and more efficient and effective responses to natural disasters. This study introduces AI-based satellite image analysis solutions throughout the disaster management cycle: prevention, preparedness, response, and recovery. Satellite imagery, captured through various channels, resolutions, and orbits, plays a crucial role throughout the entire disaster management cycle. This is because satellites have the advantage of capturing broad areas and can also image disaster regions that are inaccessible to humans due to secondary risks. We utilize high-resolution geostationary satellite imagery for real-time hazard monitoring and forecasting and synthetic Electro-Optical (EO) satellite imagery derived from Synthetic Aperture Radar (SAR) observations for monitoring flooded areas under cloudy conditions. EO satellite imagery is photographic-like images of Earth's surface using visible and infrared sensors, enabling detailed observation and analysis for applications such as mapping, surveillance, and disaster monitoring. And SAR gives high-resolution images using radar signals, capable of operating in all weather conditions and through cloud cover or darkness, making it ideal for monitoring and mapping. Additionally, AI-based damage assessment solutions facilitate the rapid detection and classification of building damage, enabling a quick response and reconstruction. Considering these technologies by government agencies, NGOs, and other stakeholders is essential, particularly in developing countries with limited surface observation capabilities and specialists. With these advanced technologies, AI-based disaster management solutions are expected to contribute significantly to the Early Warning for All initiative.

CM-UNet++: A Multi-Level Information Optimized Network for Urban Water Body Extraction from High-Resolution Remote Sensing Imagery

Urban water bodies are crucial in urban planning and flood detection, and they are susceptible to changes due to climate change and rapid urbanization. With the development of high-resolution remote sensing technology and the success of semantic segmentation using deep learning in computer vision, it is possible to extract urban water bodies from high-resolution remote sensing images. However, many urban water bodies are small, oddly shaped, silted, or spectrally similar to other objects, making their extraction extremely challenging. In this paper, we propose a neural network named CM-UNet++, a combination of the dense-skip module based on UNet++ and the CSMamba module to encode different levels’ information with interactions and then extract global and local information at each level. We use a size-weighted auxiliary loss function to balance feature maps of different levels. Additionally, features beyond RGB are incorporated into the input of the neural network to enhance the distinction between water bodies and other objects. We produced a labeled urban water extraction dataset, and experiments on this dataset show that CM-UNet++ attains 0.8781 on the IOU (intersection over union) metric, which indicates that this method outperforms other recent semantic segmentation methods and achieves better completeness, connectivity, and boundary accuracy. The proposed dense-skip module and CSMamba module significantly improve the extraction of small and spectrally indistinct water bodies. Furthermore, experiments on a public dataset confirm the method’s robustness.

Domain Adaptation and Fine-Tuning of a Deep Learning Segmentation Model of Small Agricultural Burn Area Detection Using High-Resolution Sentinel-2 Observations: A Case Study of Punjab, India

High-resolution Sentinel-2 imagery combined with a deep learning (DL) segmentation model offers a promising approach for accurate mapping of small and fragmented agricultural burn areas. Initially, the model was trained using ICNF burn area data from Portugal to capture large fire and burn area delineation, thereby achieving moderate accuracy. Subsequent fine-tuning using annotated data from Punjab improved the model’s ability to detect small burn patches, demonstrating higher accuracy than the baseline Normalized Burn Ratio (NBR) Index method. On-ground validation using buffer zone analysis and crop field images confirmed the effectiveness of DL approach. Challenges such as cloud interference, temporal gaps in satellite data, and limited reference data for training persist, but this study underscores the methodogical advancements and potential of DL models applied for small burn area detection in agricultural settings. The model achieved overall accuracy of 98.7%, a macro-F1 score of 97.6%, IoU 0.54, and a Dice coefficient of 0.64, demonstrating its capability for detailed burn area delineation. The model can capture burn area smaller than 250 m2, but the model at present is less efficient at representing the full extent of the fires. Overall, outcomes demonstrate the model’s applicability to generalize to a new domain despite regional differences among research areas.

A Small-Scale Investigation into the Viability of Detecting Canopy Damage Caused by Acantholyda posticalis Disturbance Using High-Resolution Satellite Imagery in a Managed Pinus sylvestris Stand in Central Poland

As the effects of climate change progressively worsen, many scientists are concerned over the expanding geographic range and impact of forest-defoliating insects. Many are currently pointing to this form of disturbance becoming a key focus of remote sensing research in the coming decades; however, the available body of research remains lacking. This study investigated the viability of detecting and quantifying damage caused to a managed Scots pine forest in central Poland by insect defoliation disturbance using high-resolution multispectral satellite imagery. Observed leaf area index (LAI) values were compared to frass observations (insect detritus) to assess the relationship between LAI and defoliating insect activity across a single life cycle of A. posticalis Mats. Across four managed plots, four vegetative indices (NDVI, GNDVI, EVI, and MSAVI2) were calculated using multispectral satellite imagery from a PlanetScope (PSB.SD instrument) satellite system. Then, 1137 point-sampled digital number (DN) values were extracted from each index, and a correlation analysis compared each to 40 ground-observed LAI data points. LAI was modeled on the basis of NDVI values. Three models were assessed for their performance in predicting LAI. They were fit using a variety of regression techniques and assessed using several goodness-of-fit measures. A relationship between observed LAI and frass observations was found to be statistically significant (p-value = 0.000303). NDVI was found to be the correlated LAI values (rho = 0.612). Model 3, which was based on concepts of the Beer–Lambert law, resulted in the most robust predictions of LAI. All parameters were found to be significant post fitting of the model using a nonlinear least squares method. Despite the success of the Beer’s law model in predicting LAI, detection of A. posticalis damage was not achieved. This was predominately due to issues of resolution and plot condition, among others. The results of this analysis address many interesting facets of remote sensing analysis and challenge the commonly held view of the impeachability of these methods.

Multi-Temporal and Multi-Resolution RGB UAV Surveys for Cost-Efficient Tree Species Mapping in an Afforestation Project

Accurate, cost-efficient vegetation mapping is critical for managing afforestation projects, particularly in resource-limited areas. This study used a consumer-grade RGB unmanned aerial vehicle (UAV) to evaluate the optimal spatial and temporal resolutions (leaf-off and leaf-on) for precise, economically viable tree species mapping. This study conducted in 2024 in Kasho, Bannu district, Pakistan, using UAV missions at multiple altitudes captured high-resolution RGB imagery (2, 4, and 6 cm) across three sampling plots. A Support Vector Machine (SVM) classifier with 5-fold cross-validation was assessed using accuracy, Shannon entropy, and cost–benefit analyses. The results showed that the 6 cm resolution achieved a reliable accuracy (R2 = 0.92–0.98) with broader coverage (12.3–22.2 hectares), while the 2 cm and 4 cm resolutions offered higher accuracy (R2 = 0.96–0.99) but limited coverage (4.8–14.2 hectares). The 6 cm resolution also yielded the highest benefit–cost ratio (BCR: 0.011–0.015), balancing cost-efficiency and accuracy. This study demonstrates the potential of consumer-grade UAVs for affordable, high-precision tree species mapping, while also accounting for other land cover types such as bare earth and water, supporting budget-constrained afforestation efforts.

MMRAD-Net: A Multi-Scale Model for Precise Building Extraction from High-Resolution Remote Sensing Imagery with DSM Integration

High-resolution remote sensing imagery (HRRSI) presents significant challenges for building extraction tasks due to its complex terrain structures, multi-scale features, and rich spectral and geometric information. Traditional methods often face limitations in effectively integrating multi-scale features while maintaining a balance between detailed and global semantic information. To address these challenges, this paper proposes an innovative deep learning network, Multi-Source Multi-Scale Residual Attention Network (MMRAD-Net). This model is built upon the classical encoder–decoder framework and introduces two key components: the GCN OA-SWinT Dense Module (GSTDM) and the Res DualAttention Dense Fusion Block (R-DDFB). Additionally, it incorporates Digital Surface Model (DSM) data, presenting a novel feature extraction and fusion strategy. Specifically, the model enhances building extraction accuracy and robustness through hierarchical feature modeling and a refined cross-scale fusion mechanism, while effectively preserving both detail information and global semantic relationships. Furthermore, we propose a Hybrid Loss, which combines Binary Cross-Entropy Loss (BCE Loss), Dice Loss, and an edge-sensitive term to further improve the precision of building edges and foreground reconstruction capabilities. Experiments conducted on the GF-7 and WHU datasets validate the performance of MMRAD-Net, demonstrating its superiority over traditional methods in boundary handling, detail recovery, and adaptability to complex scenes. On the GF-7 Dataset, MMRAD-Net achieved an F1-score of 91.12% and an IoU of 83.01%. On the WHU Building Dataset, the F1-score and IoU were 94.04% and 88.99%, respectively. Ablation studies and transfer learning experiments further confirm the rationality of the model design and its strong generalization ability. These results highlight that innovations in multi-source data fusion, multi-scale feature modeling, and detailed feature fusion mechanisms have enhanced the accuracy and robustness of building extraction.

Detecting Changes in Soil Fertility Properties Using Multispectral UAV Images and Machine Learning in Central Peru

Remote sensing is essential in precision agriculture as this approach provides high-resolution information on the soil’s physical and chemical parameters for detailed decision making. Globally, technologies such as remote sensing and machine learning are increasingly being used to infer these parameters. This study evaluates soil fertility changes and compares them with previous fertilization inputs using high-resolution multispectral imagery and in situ measurements. A UAV-captured image was used to predict the spatial distribution of soil parameters, generating fourteen spectral indices and a digital surface model (DSM) from 103 soil plots across 49.83 hectares. Machine learning algorithms, including classification and regression trees (CART) and random forest (RF), modeled the soil parameters (N-ppm, P-ppm, K-ppm, OM%, and EC-mS/m). The RF model outperformed others, with R2 values of 72% for N, 83% for P, 87% for K, 85% for OM, and 70% for EC in 2023. Significant spatiotemporal variations were observed between 2022 and 2023, including an increase in P (14.87 ppm) and a reduction in EC (−0.954 mS/m). High-resolution UAV imagery combined with machine learning proved highly effective for monitoring soil fertility. This approach, tailored to the Peruvian Andes, integrates spectral indices and field-collected data, offering innovative tools to optimize fertilization practices, address soil management challenges, and merge modern technology with traditional methods for sustainable agricultural practices.

Indigenous-Engaged Mapathon for Informal Road Research in Southern Siberia

ABSTRACT Extractive industries are rapidly developing in Siberian boreal forests, also known as the taiga. Expanding informal roads are undocumented, nor officially maintained, and range from wide industrial easements to narrow trails. Remote communities use informal roads for subsistence, intercommunity, and market access. This study quantifies informal road expansion surrounding Indigenous Evenki villages in Irkutsk Oblast, Russia; describes subsistence uses and ecological impacts; and offers good practice recommendations for research collaborations between Indigenous communities, researchers, and students. Interviews and mobile ethnographies informed a student-run crowdsourcing event, or “mapathon,” to digitize roads from high-resolution 1960s aerial photos and 2010s satellite imagery. Informal roads expanded more than 50 times surrounding the two villages over the last 50 years. Specific road case studies examine complex environmental justice questions regarding access, maintenance, and landscape fragmentation. Data distribution through OpenStreetMap offers empowerment through increased visibility of this unregulated development. Keywords: Indigenous knowledge, landscape fragmentation, mapathon, OpenStreetMap, Siberia