Remote Sensing Research Articles

Category-Based Interactive Attention and Perception Fusion Network for Semantic Segmentation of Remote Sensing Images

With the development of CNNs and the application of transformers, the segmentation performance of high-resolution remote sensing image semantic segmentation models has been significantly improved. However, the issue of category imbalance in remote sensing images often leads to the model’s segmentation ability being biased towards categories with more samples, resulting in suboptimal performance for categories with fewer samples. To make the network’s learning and representation capabilities more balanced across different classes, in this paper we propose a category-based interactive attention and perception fusion network (CIAPNet), where the network divides the feature space by category to ensure the fairness of learning and representation for each category. Specifically, the category grouping attention (CGA) module utilizes self-attention to reconstruct the features of each category in a grouped manner, and optimize the foreground–background relationship and its feature representation for each category through the interactive foreground–background relationship optimization (IFBRO) module therein. Additionally, we introduce a detail-aware fusion (DAF) module, which uses shallow detail features to complete the semantic information of deep features. Finally, a multi-scale representation (MSR) module is deployed for each class in the CGA and DAF modules to enhance the description capability of different scale information for each category. Our proposed CIAPNet achieves mIoUs of 54.44%, 85.71%, and 87.88% on the LoveDA urban–rural dataset, and the International Society for Photogrammetry and Remote Sensing (ISPRS) Vaihingen and Potsdam urban datasets, respectively. Compared with current popular methods, our network not only achieves excellent performance but also demonstrates outstanding class balance.

Improved Faster R-CNN Based on Multi-Scale Module for Detecting Weak Ground Targets in Remote Sensing Images

In order to optimize the detection capability of small and weak ground targets in remote sensing images, a remote sensing image ground weak target detection method based on multi-scale module improved Faster R-CNN is proposed. By adjusting the Region Proposal Network (RPN) and Region of Interest (RoI) pooling layers, the model is better suited for feature extraction and classification of small targets. Generate feature maps at multiple scales by introducing additional convolutional layers to process inputs at different resolutions, with each layer focusing on targets within a specific size range. These multi-scale features are integrated and input into RPN to provide richer target proposal characteristics. By using image transformation techniques such as scaling, cropping, and rotation, the training dataset is expanded to simulate different small target scenes, enhancing the model’s adaptability to changes in target size, direction, and environment. The experimental results show that the proposed method not only outperforms other models in accuracy, recall, and F1 score, but also maintains high consistency and stability across different datasets. Its significant advantage in mean accuracy (MAP) further proves its reliability and effectiveness in remote sensing applications.

Multi-Scale Expression of Coastal Landform in Remote Sensing Images Considering Texture Features

Multi-Scale Expression of Coastal Landform in Remote Sensing Images Considering Texture Features

Instance-Level Scaling and Dynamic Margin-Alignment Knowledge Distillation for Remote Sensing Image Scene Classification

Instance-Level Scaling and Dynamic Margin-Alignment Knowledge Distillation for Remote Sensing Image Scene Classification

Multi-Source Remote Sensing Data for Wetland Information Extraction: A Case Study of the Nanweng River National Wetland Reserve

Wetlands play a vital role in regulating the global carbon cycle, providing biodiversity, and reducing flood risks. These functions maintain ecological balance and ensure human well-being. Timely, accurate monitoring of wetlands is essential, not only for conservation efforts, but also for achieving Sustainable Development Goals (SDGs). In this study, we combined Sentinel-1/2 images, terrain data, and field observation data collected in 2020 to better understand wetland distribution. A total of 22 feature variables were extracted from multi-source data, including spectral bands, spectral indices (especially red edge indices), terrain features, and radar features. To avoid high correlations between variables and reduce data redundancy, we selected a subset of features based on recursive feature elimination (RFE) and Pearson correlation analysis methods. We adopted the random forest (RF) method to construct six wetland delineation schemes and incorporated multiple types of characteristic variables. These variables were based on remote sensing image pixels and objects. Combining red-edge features, terrain data, and radar data significantly improved the accuracy of land cover information extracted in low-mountain and hilly areas. Moreover, the accuracy of object-oriented schemes surpassed that of pixel-level methods when applied to wetland classification. Among the three pixel-based schemes, the addition of terrain and radar data increased the overall classification accuracy by 7.26%. In the object-based schemes, the inclusion of radar and terrain data improved classification accuracy by 4.34%. The object-based classification method achieved the best results for swamps, water bodies, and built-up land, with relative accuracies of 96.00%, 90.91%, and 96.67%, respectively. Even higher accuracies were observed in the pixel-based schemes for marshes, forests, and bare land, with relative accuracies of 98.67%, 97.53%, and 80.00%, respectively. This study’s methodology can provide valuable reference information for wetland data extraction research and can be applied to a wide range of future research studies.

Four Decades of Polar Research in Remote Sensing: A Comprehensive Review

Four Decades of Polar Research in Remote Sensing: A Comprehensive Review

National Spatial Data Policy and Remote Sensing: Technological Advancements, Policy Implications, and Future Prospects

National Spatial Data Policy and Remote Sensing: Technological Advancements, Policy Implications, and Future Prospects

A Comprehensive Review of Remote Sensing for Water-Related Disaster Management in South Korea: Focus on Floods and Droughts

A Comprehensive Review of Remote Sensing for Water-Related Disaster Management in South Korea: Focus on Floods and Droughts

History, Status, and Prospects of Remote Sensing in Agriculture in Republic of Korea

History, Status, and Prospects of Remote Sensing in Agriculture in Republic of Korea

History, Status, and Prospects of Remote Sensing in Korea

History, Status, and Prospects of Remote Sensing in Korea

Advancement and Applications of Forest Remote Sensing in Korea: Past, Present, and Future Perspectives

Advancement and Applications of Forest Remote Sensing in Korea: Past, Present, and Future Perspectives

History, Status, and Prospects of Remote Sensing in the Field of Meteorological Satellite in Korea

History, Status, and Prospects of Remote Sensing in the Field of Meteorological Satellite in Korea

Quantifying NO x Emission Sources in Houston, Texas Using Remote Sensing Aircraft Measurements and Source Apportionment Regression Models.

Air quality managers in areas exceeding air pollution standards are motivated to understand where there are further opportunities to reduce NO x emissions to improve ozone and PM2.5 air quality. In this project, we use a combination of aircraft remote sensing (i.e., GCAS), source apportionment models (i.e., CAMx), and regression models to investigate NO x emissions from individual source-sectors in Houston, TX. In prior work, GCAS column NO2 was shown to be close to the "truth" for validating column NO2 in model simulations. Column NO2 from CAMx was substantially low biased compared to Pandora (-20%) and GCAS measurements (-31%), suggesting an underestimate of local NO x emissions. We applied a flux divergence method to the GCAS and CAMx data to distinguish the linear shape of major highways and identify NO2 underestimates at highway locations. Using a multiple linear regression (MLR) model, we isolated on-road, railyard, and "other" NO x emissions as the likeliest cause of this low bias, and simultaneously identified a potential overestimate of shipping NO x emissions. Based on the MLR, we modified on-road and shipping NO x emissions in a new CAMx simulation and increased the background NO2, and better agreement was found with GCAS measurements: bias improved from -31% to -10% and r2 improved from 0.78 to 0.80. This study outlines how remote sensing data, including fine spatial information from newer geostationary instruments, can be used in concert with chemical transport models to provide actionable information for air quality managers to identify further opportunities to reduce NO x emissions.

Integrated Morphometric and Hypsometric Analysis of Niragantipalli Micro Watershed Using Remote Sensing and Geographical Information System Techniques

Employing Remote sensing (RS) and geographic information system (GIS) for the morphometric parameters analysis are discovered to be tremendous usefulness in the prioritization of watersheds for soil, water conservation and natural resource management at micro watershed level. The analysis of morphometric parameters plays a crucial role for understanding and managing watersheds. In current study an, attempt was to determine the morphological parameters of Niragantipalli micro-watershed in Chikkaballapura District of Karnataka, India. For detail study, Google earth pro, GPS visualizer conversion toll and Arc GIS were used for preparation of DEM and delineation of the watershed boundary. GIS was used for evaluation of basic, linear, areal and relief aspects of morphometric parameters. The Niragantipalli micro-watershed occupies an area of 632 ha with third order stream (truck order) and dendritic drainage pattern. The mean bifurcation ratio of the micro-watershed is 2.5 it suggests that the drainage system formed on homogeneous rock when the influences of geologic structures on the network of streams were negligible and generate sharp peak flow. With a drainage density of 1.31 km/km2, the drainage network is extremely coarse. The watershed is elongated, as shown by the form factor of 0.27 and elongation ratio of 0.58. The results from the morphometric analysis of the watershed are very beneficial for developing and designing conservation structures of soil and watershed management measures. The hypsometric curves’ structure as well as estimated hypsometric integral results reflects the Niragantipalli Micro Watershed erosional stages. As a result, the study concludes that morphometric and hypsometric analysis findings may be useful to stakeholders participating in catchment development and management projects.

Groundwater potential and recharge zone mapping using GIS and remote sensing techniques: the Melka Kunture Watershed in Ethiopia

Groundwater is the most dynamic natural resource that is not uniformly distributed both in space and time. Identifying it for sustainable water resource development for domestic use, irrigation, and industrial purposes is the biggest challenge and key concern due to climate change, overexploitation, and a lack of proper management. Geographical information system (GIS) and remote sensing (RS) play a fundamental role in identifying suitable groundwater potential and recharge zones. In this study, groundwater potential and recharge zone map are delineated using GIS and RS techniques integrated with Analytic Hierarchy Process (AHP) for the Melka Kunture Watershed in Ethiopia. The thematic layers used were: Lithology, geomorphology, slope, lineament density, land use/cover, rainfall, drainage density and soil. GIS has been used to integrate and overlay thematic layers using a weighted overlay analysis tool, whereas the AHP method has been used to assign the weights, normalize, and rank thematic layers and their associated sub-themes based on their suitability, characteristics, or influence on groundwater potential and recharge zones. The output groundwater potential zone map is categorized into four suitable zones: high potential cover 1105.60 km2 (25.70%), moderate potential 2308.00 km2 (53.73%), low potential 880.0 km2 (20.49%), and very low potential 47.80 km2 (1.10%), whereas there are three suitable groundwater recharge potential zones: high potential cover 250.20 km2 (5.80%), moderate recharge 3647.90 km2 (84.00%), and low recharge 443.50 km2 (10.20%) of the study area. Hence, the study areas have been more suitable for groundwater potential and recharge zones as compared to the total area. The groundwater potential map was validated using the existing water source and well points, and it indicated a good prediction accuracy of 88%. Thus, the delineated groundwater potential and recharge zone maps are reliable, and the resultant potentiality mapping has played a vital role in the sustainable development and management of the water resources in the study area.

Evaluation of Potential Flood Areas in the Basin of Lake Ladik through AHP and GIS Integration, (Samsun, Türkiye)

Floods are a problem of many countries on a global scale. In Türkiye, especially in the summer months, large floods occur in the Black Sea Region coastal belt. The number and the destructive power of experienced floods are increasing day by day. It is not possible to prevent the occurrence of floods. But it is possible to take the necessary measures to prevent it from turning into a disaster. Recently, susceptibility analyses have been carried out for floods and similar natural disasters and the results help decision-makers. Within the scope of this study, flood susceptibility analysis was carried out by using multiple geographical factors together in the Basin of Lake Ladik of Samsun Province in the Black Sea Region. In the study, Analytical Hierarchy Process (AHP), Geographic Information Systems (GIS) and Remote Sensing (RS) techniques were used, which are multi-criteria decision-making methods. In this context, flood susceptibility analysis of the basin was carried out by using nine (9) different geographical factors (slope, aspect, lithology, soil, basin size, land use, landforms, precipitation, and drainage density). As a result of the study, for floods, there were identified four (4) different levels as low, medium, high, and very high; and 36.77% of the basin was found to be low, 30.03% was medium, 11.43% was high and 21.77% was found to be sensitive to possible floods at a very high level. The results of the study are also important for decision-makers to make flood risk planning.

Analysis of the Morphometric Characteristics of Wadi Rajil in the Azraq Basin using Geographic Information Systems

The study aimed to Analyse the Hydrometric Characteristics of Wadi Rajil basin and assess the flood hazards in the basin using Remote Sensing and Geographic Information Systems Techniques. The study used Models of Digital Visualizations of the basin and processed them using ARC GIS 10.5 and ARC HYDRO TOOLS. The result was that the Hypsometric Coefficient reached (21.4%) of the valley area that was exposed to erosion and (78.6%) of it is still waiting for its turn to erosion, indicating that the valley has reached the stage of youth. The analysis of Hot Spots in the GIS program resulted in negative values for internal Drainage Basins greater than -0.5 extending to the northern and northwestern parts of the basin. It also showed that the flow coefficient exceeds the internal Drainage Coefficient or Leakage. As for the positive values, they are distributed in the southern and southeastern parts, where the (Hot) values prevail and where the leakage coefficient exceeds the Flow Coefficient. We find that the Hot Spots are concentrated in the southeastern parts, which are mostly extended. The Rift and thus the most suitable areas for Groundwater Charging through it.

Unveiling the outcome of multispectral indices in evaluating total suspended solid of water quality

Constant monitoring of TSS in water body remains essential to avoid excessive suspended sediment that can damage water quality. This study examines TSS as a visible indicator of water quality parameter in Hartbeespoort Dam located in the Northwest of South Africa. The study employed multi-spectral spatial bands of Landsat 8–9 (OLI & TIR) and MSI of Sentinel-2A and 2B. Band combination indices such as NDSSI, WRSI was utilized respectively using Landsat 8–9 and MSI Sentinel 2. Nechad TSS algorithm and Automated Water Extraction Index (AWEI) was also utilized with green band and red band of Landsat 8–9 and Sentinel 2 to further understand the concentration of suspended solid. The values derived was correlated with real time in-situ data measurement. NDSSI mapping of TSS showed high values range of 0.895 (suspended solid) while WRSI in Sentinel 2 spatial band showed highest value of 0.551 (suspended solid). The regression analysis utilizes Linear, Exponential and polynomial equation to validate with the in-suit data measurement. The result revealed a strong correlation such as (NDSSI and in-suit =(R2 = 0.54) in polynomial equation, WRSI and in-suit = (R2 = 0.71) in exponential regression, Nechad algorithm and in-suit = (R2 = 0.54) in polynomial regression and AWEI with in-suit = (R2 = 0.71) in exponential regression. The value derived from Nechad TSS algorithm and AWEI confirmed the capacity of NDSSI indices in Landsat 8–9 and WRSI in Sentinel 2 remote sensor.

Creating Sustainable Flood Maps Using Machine Learning and Free Remote Sensing Data in Unmapped Areas

This study leverages a Random Forest model to predict flood hazard in Arizona, New Mexico, Colorado, and Utah, focusing on enhancing sustainability in flood management. Utilizing the National Flood Hazard Layer (NFHL), an intricate flood map of Arizona was generated, with the Random Forest Classification algorithm assessing flood hazard for each grid cell. Weather variable predictions from TerraClimate were integrated with NFHL classifications and Digital Elevation Model (DEM) analyses, providing a comprehensive understanding of flood dynamics. The research highlights the model’s capability to predict flood hazard in areas lacking NFHL classifications, thereby supporting sustainable flood management by elucidating weather’s influence on flood hazard. This approach aligns with sustainable development goals by aiding in resilient infrastructure design and informed urban planning, reducing the impact of floods on communities. Despite recognizing constraints such as input data precision and the model’s potential limitations in capturing complex variable interactions, the methodology offers a robust framework for flood hazard evaluation in other regions. Integrating diverse data sources, this study presents a valuable tool for decision-makers, supporting sustainable practices, and enhancing the resilience of vulnerable regions against flood hazards. This integrated approach underscores the potential of advanced modeling techniques in promoting sustainability in environmental hazard management.

Multi-feature fusion and multi-attention deep network for enhancing road extraction in remote sensing images

ABSTRACT Road detection in remote sensing (RS) images plays a critical role in applications ranging from urban planning to autonomous navigation systems. However, accurate road extraction remains a challenging task due to the presence of textual-similar objects that can be visually confused with roads, and shadows that can obscure road features. For this regard, we present a novel end-to-end network that leverages multi-feature fusion and multi-attention. Multi-level dual residual blocks are introduced to capture multi-scale and multi-level road features. A channel attention feature fusion module fuses road features from the decoder while suppressing coarse-grain noises. Moreover, a difference channel feature fusion module filters out the interfering texture-similar features and irrelevant background features. Subsequently, a spatial channel feature enhancement module is designed to identify the mis-segmented regions and reclassify these pixels. We conducted extensive experiments on diverse datasets, demonstrating the effectiveness of our approach in improving road detection accuracy, especially in areas with shadows and textual-similar objects. The results indicate that our approach outperforms state-of-the-art methods in handling textual-similar objects and shadow-occluded roads, making it a valuable contribution to the field of road extraction in remote sensing images.