最新刊期
- “在遥感图像语义分割领域，FLGF-UNet网络有效克服漏检、误检等问题，为城市变化检测等提供解决方案。”
摘要：Semantic segmentation of remote sensing images plays a crucial role in fields such as urban change detection, environmental protection, and geological disaster identification. Addressing issues in current remote sensing building extraction, such as missed detections, false positives, and incomplete extractions due to tree obstruction or similar object interference, this paper proposes an improved building extraction network based on the UNet architecture - the Fusion of Local Global Features Network (FLGF-UNet). This model integrates local and global information, acquiring features in parallel through multi-scale local and global modeling. This ensures that the feature representation at each stage of the network incorporates both local and global information, enabling a more natural and comprehensive integration of information across different scales and levels during the feature extraction process.The parallel feature fusion method of FLGF-UNet ensures that the features of each stage contain fine-grained local information and global dependencies, so that the network has both local and global information in the feature representation of each stage, effectively overcoming the shortcomings of Transformer in local information exchange, and at the same time outperforming traditional CNN in global information modeling. The LF module is introduced to extract local information and global context information of different scales to ensure the integrity of feature dependencies, thereby making up for the shortcomings of a single module in feature extraction. In addition, in order to bridge the semantic gap between the encoder and decoder, an interactive fusion (IF) module is added between the encoder and decoder to enhance the fusion effect of spatial details, global context and semantic features. To verify the superiority and versatility of FLGF-UNet, the proposed network is compared with networks such as U2Net, Swin Transformer, MA-Net, HD-Net and RS-Mamba on the WHU, Massachusetts datasets and typical urban building instance datasets in China. The results show that FLGF-UNet outperforms other SOTA networks in performance and has high practical application value.In conclusion, FLGF-UNet is an innovative network for extracting buildings from high-resolution remote sensing images. It takes parallel multi-scale local-global modeling as its core, so that features at all levels can simultaneously have local details and long-range semantics, effectively bridging the gap between spatial dependence and local details, and significantly improving extraction accuracy. Extensive experiments across data sets have verified that its performance is significantly better than existing methods, providing a reliable solution for high-precision building extraction from high-resolution remote sensing images. FLGF-UNet brings together cutting-edge methods and technological innovations, marking an important leap forward in remote sensing image analysis, and will continue to drive the in-depth development of this field in terms of accuracy improvement, scene expansion, and practical applications.In the future, the breakthrough results of FLGF-UNet will inspire the academic community to continue to delve deeper into the local-global structure. Migrating and expanding this fusion strategy to more remote sensing interpretation tasks has great potential and promising prospects. At the same time, in view of the diversity of urban scenes in scale and complexity, further optimizing the model architecture and improving adaptive capabilities are also a natural evolutionary path. Subsequent research should focus on the implementation of technology and effectively transform the advantages of the algorithm into perceptible, quantifiable, and sustainable socioeconomic benefits for urban planners and disaster emergency decision makers.关键词：remote sensing images;building extraction;fusion of local-global feature networks;Feature fusion;interactive fusion module227|111|0更新时间：2026-01-15 - “在海洋环境监测领域，专家建立了海浪风场参数联合反演模型，为中高海况下海浪风场参数高精度反演提供解决方案。”
摘要：Objective Synthetic Aperture Radar (SAR), as an active microwave sensor, has high resolution and the ability to observe the sea in all-weather and all-time conditions, playing an important role in monitoring ocean wave and wind fields. However, existing wave and wind field parameter inversion methods still face low inversion accuracy in moderate to high sea states. In this work, we established two joint inversion models of wave and wind field parameters suitable for moderate to high sea states and verified their inversion performance, providing an important reference for subsequent marine environmental studies under extreme sea conditions.Method Addressing the problem of low inversion accuracy of existing wave wind field parameter inversion methods under medium to high sea conditions, this paper utilizes Sentinel-1 WV mode SAR data and introduces three approximate polarization feature parameters (polarization entropy, anisotropy, and correlation) as additions based on traditional characteristic parameters (incident angle, normalized variance). Meanwhile, based on Support Vector Regression(SVR) and Convolutional Neural Network(CNN) algorithms, two joint inversion models for SAR wave wind field parameters suitable for medium to high sea conditions (Wind speed is 10.8 m/s - 28.8 m/s)were constructed.Result The results show that the root mean square errors (RMSE) of wind speed, wind direction, significant wave height, and average wave period inverted based on SVR, compared to ERA5 and NDBC data, are 1.27m/s, 23.46°, 0.22m, 0.62s and 1.10m/s, 25.37°, 0.22m, 0.59s respectively; the RMSE for wind speed, wind direction, significant wave height, and average wave period inverted based on CNN compared to ERA5 and NDBC data are 1.15m/s, 23.53°, 0.18m, 0.53s and 1.08m/s, 25.85°, 0.19m, 0.54s, demonstrating that both joint inversion models established in this paper can effectively invert wave wind field parameters under medium to high sea conditions. Compared to traditional theoretical methods, the joint inversion models based on SVR and CNN significantly reduced the RMSE of significant wave height, average wave period, and wind speed compared to ERA5 data by 0.67m, 0.62s, 0.3m/s and 0.71m, 0.71s, 0.42m/s respectively, proving that both joint inversion models significantly enhance inversion accuracy. Furthermore, to verify the inversion effect of the two joint inversion models under high sea conditions, two scenes of hurricane data from the North Atlantic in 2020 were collected for case analysis. The results indicate that both joint inversion models established in this paper can also achieve relatively accurate wave wind field parameters under high sea conditions.Conclusion In conclusion, the joint inversion models developed in this paper provide an effective solution for high-precision inversion of wave wind field parameters under medium to high sea conditions and offer important references for subsequent marine environment studies under extreme sea conditions.关键词：synthetic aperture radar;joint inversion method;CNN;SVR;wave wind field parameters;moderate to high sea conditions;polarization characteristic parameters125|164|0更新时间：2026-01-15 - “在岩溶塌陷监测领域，研究者提出了改进同质像元识别算法，有效提升了DS-InSAR技术监测精度和稳定性，为深入理解死海地区岩溶塌陷灾害提供了重要数据支撑与理论参考。”
摘要：Objective Karst sinkholes represent a significant geohazard characterized by complex spatiotemporal evolution and causative mechanisms. This study aims to enhance the applicability of Distributed Scatterer InSAR (DS-InSAR) in low-coherence karst terrains and to clarify the hydrological controls governing sinkhole development along the Dead Sea.Method To address the limitations of inaccurate homogeneous pixel selection and unstable phase estimation, we propose a dynamic-confidence-interval algorithm for homogeneous-pixel identification (D-HTCI). This method iteratively updates the reference mean and confidence interval per pixel to mitigate estimation biases. Taking the Dead Sea karst region as a case study, we integrated D-HTCI with a sequential phase-linking strategy to reconstruct a continuous, high-density deformation field using 242 Sentinel-1A scenes acquired between 2016 and 2024.Results The proposed approach retrieved 832,000 monitoring points, representing a substantial increase of 367,000 and 153,000 points compared to PS-InSAR and conventional DS-InSAR, respectively. Deformation rates primarily ranged from -120 to 20 mm/yr, with maximum cumulative subsidence exceeding 800 mm in the southwestern collapse zone. Time-series analysis revealed a strong linear coupling with synchronous Dead Sea water-level changes (R² > 0.98), quantitatively supporting a hydrogeological driver whereby sustained water-level decline lowers the fresh–saline interface, promotes salt dissolution, and triggers collapse.Conclusion The integration of D-HTCI with sequential phase optimization significantly improves the feasibility and accuracy of DS-InSAR for monitoring natural terrains. This study provides compelling quantitative evidence and a robust framework for understanding the spatiotemporal evolution of karst sinkholes.关键词：Karst collapse;Dynamic Confidence Intervals;Homogeneous pixel identification;DS-InSAR;deformation monitoring;Water level decline173|78|0更新时间：2026-01-15 - “在广域地质灾害监测领域，我国首个L波段干涉测量卫星星座陆地探测1号卫星数据处理方法取得新进展，有效提升了卫星数据的应用价值。”
摘要：Objective The LuTan-1 (LT-1) satellite, China’s first L-band interferometric SAR constellation, offers significant advantages for large-scale geological hazard monitoring due to its long wavelength and high spatiotemporal resolution. However, early-stage limitations such as insufficient real-time orbit accuracy, irregular image frames, and low effective image overlap rates significantly constrain its potential for wide-area applications. This study aims to develop a wide-area LT-1 data processing strategy that explicitly accounts for image overlap rate, thereby improving both processing efficiency and deformation monitoring accuracy.Method The proposed approach classifies images by orbital path and, within each path, constructs interferometric networks optimized by constraints on both spatio-temporal baselines and overlap ratios. A combination of interferogram stacking (Stacking) and regional network adjustment is employed to achieve seamless mosaicking of deformation results from different paths. On this basis, for key areas, the Small Baseline Subset (SBAS-InSAR) technique is applied to perform refined time-series deformation inversion. The Chongqing section of the Three Gorges Reservoir Area (including Yunyang, Fengjie, and Wushan counties) serves as the experimental area. The methodology is evaluated against a conventional LT-1 processing workflow to quantify improvements in data utilization and deformation accuracy.Result Compared with the traditional frame-based network construction strategy, the proposed method improves image utilization by 62% and achieves a deformation monitoring accuracy of 6.1 mm/yr, outperforming the 10.7 mm/yr obtained using the conventional frame-based network construction approach . Landslide detection results indicate that LT-1 ascending and descending track data identify 57 and 63 more potential landslide hazard sites, respectively, than Sentinel-1 ascending track data. The findings highlight the distinctive capability of L-band SAR satellites for landslide detection in mountainous terrain with complex surface conditions.Conclusion The proposed wide-area LT-1 data processing method effectively addresses challenges of low efficiency and reduced accuracy caused by irregular image coverage and low overlap rates. It significantly enhances LT-1’s applicability for large-scale geological hazard monitoring and demonstrates superior landslide detection performance over C-band systems in complex terrain. As LT-1 data archives grow and orbital accuracy improves, this method is expected to further increase the precision and reliability of deformation monitoring for wide-area hazard assessment.关键词：wide-area InSAR;Lutan-1;overlap-rate constraint;deformation monitoring;landslide detection;Three Gorges Reservoir Area;Stacking;L-band365|260|0更新时间：2025-12-11 - “在城市规划领域，研究者基于开源数据和遥感影像，提出了一种自动识别混合功能街区的方法，有效提升了城市功能多样性识别的精度。”
摘要：Objective The identification of the functions of urban blocks serves as a crucial foundation for urban planning and management. With the acceleration of urbanization, the division of single-functional zones can no longer adequately meet the demands of complex urban spaces. As a manifestation of multifunctional urban integration, the identification of mixed-function blocks—particularly automated identification—holds significant importance for understanding urban functional diversity and enhancing land use efficiency. Against this backdrop, this study proposes an automated sample extraction method for seven categories, including both pure and mixed use, by integrating single functional information derived from Area of Interest (AOI) and Point of Interest (POI) data, along with OpenStreetMap (OSM) and Sentinel-2 imagery. The ResNet34 model is then employed to achieve functional identification for each street block.Method First, the information entropy of POI distribution is utilized to distinguish between single functional and mixed use street blocks, forming the initial sample set. Subsequently, a multi-view discrepancy learning module, based on Sentinel-2 imagery and single functional samples, is designed to further extract samples for both single-used and mixed-use categories. Considering the scale discrepancy between AOIs and actual urban blocks, the above mentioned automated sample extraction scheme is applied to both AOI and street block units to enhance sample quantity and scale diversity.Result The proposed automatic classification method in this study achieved overall accuracies (OAs) of 72.9%, 78.3%, 73.4%, and 75.1% in Beijing, Hefei, Weifang, and Chengdu, respectively. Compared to the approach using solely POI distribution entropy, the combined use of AOI and POI data improved the recognition accuracy for mixed-function categories by 7%, 18%, 20%, and 13% in these four cities.Conclusion These results demonstrate the feasibility and effectiveness of the method across diverse urban environments, as well as the potential of integrating crowdsourced geographic data and remote sensing imagery in urban functional zone studies—particularly in the context of mixed use urban functional zones.关键词：mixed-use street block;Sentinel-2 imagery;deep learning;POI;AOI;urban function zone;multi-view learning223|327|0更新时间：2025-12-11 - “新疆三道坝煤田火区监测研究取得新进展，专家构建了基于STL时序分解的煤火监测方法，提高了监测精度，为煤火治理提供参考。”
摘要：Coal fires represent a major global environmental hazard characterized by long combustion cycles, strong concealment, and considerable difficulty in mitigation. They pose severe risks to ecological security, human health, and energy resources. As coal-fire evolution is a continuous spatiotemporal process, Land Surface Temperature (LST) serves as a key indicator for identifying thermal anomalies and tracking fire development. With the rapid accumulation of multi-source remote sensing data, time-series analysis has become a powerful tool for long-term coal-fire monitoring. However, LST data often exhibit strong randomness and complex variability, which introduces challenges for long-term sequence analysis. This study aims to develop a robust coal-fire monitoring method capable of accurately characterizing long-term LST variations and identifying coal-fire evolution patterns. The Sandaoba coalfield in Xinjiang, China, was selected as the study area. Using Landsat imagery and the Google Earth Engine (GEE) platform, a continuous LST dataset from 1998 to 2023 was constructed. The Seasonal-Trend decomposition procedure based on Loess (STL) was applied to separate the LST time series into trend, seasonal, and residual components, allowing the removal of strong seasonal fluctuations and stochastic disturbances. Spatiotemporal patterns of LST were analyzed using the extracted trend component. To further identify coal-fire development stages and delineate active fire zones, the Random Sample Consensus (RANSAC) algorithm was used to fit long-term temperature trends at the pixel scale. Field survey data collected in 2016 were employed to validate the identification results. we obtained the following results: The STL decomposition effectively separated the long-term LST series into stable trend components and variable seasonal and residual signals. The extracted trend item revealed long-term warming trajectories associated with coal-fire activity more accurately compared with the raw LST series. Among the 20 coal-fire points detected in the 2016 field survey, 16 fell within high-value areas of the mean and range of the trend component, indicating strong spatial consistency. The RANSAC based trend fitting captured spatiotemporal LST evolution from 1998 to 2023 and demonstrated high agreement with field observations, successfully identifying the initiation, expansion, and gradual stabilization stages of coal fires. These results validate the robustness and reliability of combining STL decomposition with RANSAC for long-term coal-fire monitoring. The proposed STL-based method significantly improves the capability to extract meaningful long-term LST trends and enhances the adaptability of coal-fire monitoring to complex spatiotemporal conditions. The method accurately identifies thermal anomalies associated with coal-fire development and shows strong consistency with field investigation results. This framework enables long-term, large-scale LST analysis and provides an effective tool for capturing the spatiotemporal characteristics of coal-fire evolution. The findings offer valuable technical support for future coal-fire surveillance, early warning, and mitigation planning.关键词：Coal fire identification;STL decomposition;Landsat;LST time series;thermal infrared remote sensing309|220|0更新时间：2025-12-11 - “最新研究利用全散射体InSAR技术，精细监测盐穴型储气库地表形变，为保障储气库安全稳定运行提供重要参考。”
摘要：Objective China faces significant contradictions in natural gas supply-demand and substantial peak-shaving requirements, making underground gas storage facilities critical infrastructure for ensuring stable gas supply. However, salt rock creep induced by high-pressure operations in salt cavern gas storage can trigger pronounced ground deformation, posing severe threats to the safe production of storage facilities. Accurately mapping fine ground deformation of gas storage has thus become a prerequisite for ensuring their safe and stable operation. This study aims to characterize the spatio-temporal patterns of ground deformation in a large-scale salt cavern gas storage using advanced time series interferometric synthetic aperture radar (InSAR) technique, and to validate the feasibility of high-precision monitoring for hazard prevention.Method This research employs the latest Full Scatterer InSAR (FS-InSAR) technique, with the ability of separating temporal low-frequency deformation phase from temporal high-frequency phases such as atmospheric delay contribution, to process 79 Sentinel-1 C-band SAR images acquired from August 2021 to August 2024. The dataset covers a large-scale salt cavern gas storage in eastern China. Through time-series SAR processing, including interferogram generation, phase unwrapping, dual-scale temporal low-pass filtering and deformation inversion, the study retrieves millimetric-level ground deformation rates and cumulative deformation. Additionally, leveling measurements collected synchronously are used to validate the accuracy of FS-InSAR results.Result The FS-InSAR technology can obtain high-precision and high-density ground deformation in gas storage areas with abundant water and farmland. The monitoring density reaches up to 1,121points/km², enabling full-pixel deformation monitoring for all areas except water bodies. Validation by 28 synchronously levelling data shows that the standard deviation of the difference between the FS-InSAR and leveling measurements is 2.9 mm/a.Conclusion This study highlights the effectiveness of FS-InSAR in providing high-resolution, continuous monitoring of surface deformation in salt cavern gas storage facilities. The identified spatial heterogeneity and periodic deformation patterns underscore the importance of integrating radar remote sensing with operational data for real-time safety assessment. The technology enables early detection of high-risk zones (e.g., old caverns with severe subsidence) and facilitates adaptive management of gas storage operations. The findings contribute to the development of intelligent monitoring systems for underground energy storage infrastructure, supporting sustainable and secure gas supply in China’s energy transition.关键词：InSAR;full scatterer (FS);salt cavern gas storage;surface deformation;gas injection and production448|225|0更新时间：2025-12-11 - “在地形断裂线提取领域，专家提出了一种新算法，有效提升了断裂线的完整性和正确性，为数字高程模型建模提供了新方案。”
摘要：(Objective)Topographic break lines are key structural lines that describe complex terrain features and optimize terrain representation. They can precisely depict the geometric shapes and structural details of the terrain. Accurate extraction of these lines is of great significance for enhancing the authenticity and reliability of digital terrain models. However, accurately and completely extracting topographic break lines from point cloud data has always been a core technical challenge in the field of digital terrain modeling.(Method)In response to the common problems such as low completeness, insufficient accuracy, and severe over- and under-extraction in current methods for extracting topographic break lines based on ground point clouds, this paper proposes a topographic break line extraction algorithm that takes into account multi-scale characteristics and anti-shrinkage fracture. The aim is to significantly improve the accuracy and robustness of topographic break line extraction. This method first addresses the issue that single-scale characteristics are insufficient to comprehensively capture the complex terrain changes by using a random forest model based on multi-scale and multi-feature to capture potential terrain feature points. Then, by analyzing the point cloud contraction trend, potential feature points are classified into potential ridge and valley feature points, and regional growth clustering with principal direction consistency constraints is used for denoising. Subsequently, to address the problem of feature lines breaking and endpoint contraction caused by traditional Laplacian smoothing, the Laplacian with vertical constraints is used to smooth and refine the denoised potential ridges and valleys feature points. Finally, under the constraint of connecting edge clipping rules to eliminate false connections, a minimum spanning tree is constructed to obtain high-quality topographic break lines.(Result)To verify the validity and superiority of the method, experiments were conducted in two typical terrains: mining areas and mountainous regions, and quantitative and qualitative comparisons were made with three mainstream methods (LapS, D8, and PIM). The results showed that the method proposed in this paper was significantly superior to the comparison methods in terms of completeness, accuracy, and comprehensive quality indicators. In each complex terrain area, the completeness, accuracy, and quality of topographic break line extraction were at least improved by 10.4%, 5.8%, and 11.8% respectively. In addition, the study also evaluated the accuracy of constructing a digital elevation model by using the proposed topographic break lines as constraints. The experiments found that, under different point cloud densities, especially when the data density was lower than 10%, the DEM accuracy constrained by the topographic break lines was significantly better than that without constraints. The root mean square error (RMSE) and mean absolute error (MAE) of the DEM increased by 55.9% to 82.9%, effectively maintaining the terrain details and overall shape.(Conclusion)In conclusion, the method proposed in this paper, taking into account multi-scale characteristics and anti-shrinkage fracture strategies, significantly improves the accuracy and completeness of topographic break line extraction. On this basis, it effectively supports the construction of high-fidelity digital elevation models, thereby providing a reliable technical solution for terrain feature analysis and high-precision digital terrain modeling in complex environments.关键词：airborne LiDAR;Topographic break lines;Multi-scale terrain characteristics;Laplacian smoothing with constraints;digital elevation model71|121|0更新时间：2025-12-11 - “在遥感蒸散模型验证领域，专家利用光学-微波双波长光闪烁方法，验证了与卫星观测时空尺度相匹配的地面观测数据，为ET模型验证提供了有效途径。”
摘要：Most current thermal-based remote sensing evapotranspiration (ET) models estimate sensible heat flux (H), latent heat flux (LE), and ET using surface temperature and other parameters derived from multispectral and thermal imagery acquired within tens of seconds during the satellite overpasses. However, their instantaneous flux estimates are commonly validated against 30-min averaged flux data from the eddy-covariance (EC) method, leading to a significant temporal scale mismatch. Developed over the past two decades, optical-microwave two-wavelength scintillometry measures H and LE at scales ranging from hundreds of meters to 10 km, is suitable for complex environments including mountainous terrain, and yields statistically stable flux results with a shorter averaging time of 1~ 2 min. As a complement to EC, this method provides a critical opportunity for validating remote sensing ET models and products. Based on observations from optical-microwave scintillometer (OMS) and EC systems as well as meteorological profile towers in alpine grasslands of the upper reaches and oasis croplands of the middle reaches of the Heihe River Basin (HRB), northwest China, this study highlights the outstanding advantages of OMS flux observations in temporal-scale matching with ET model estimates. Comparative analysis of OMS and EC flux across different averaging times (1, 2, 5, 10, 15, 30 min) confirms the theoretical feasibility that OMS can generate statistically stable fluxes with averaging periods as short as 1 min. This time scale is highly consistent with the scanning duration of single satellite images (e.g., Landsat). Based on this finding, analysis of 1-min OMS data in alpine grassland shows that H and LE over the 30-min periods typically used for model validation often vary by 10~30%. Using the 1-min flux data at the time of satellite overpass to validate ET model results introduces uncertainties of similar magnitude, which are exacerbated under atmospheric nonstationary conditions. Validation of instantaneous H and LE estimated by the Two-source surface energy balance (TSEB) model and Landsat images using OMS and EC flux data from oasis croplands demonstrates better agreement between TSEB outputs and 1-min OMS flux observations. Optical-microwave scintillometry enables ground observations to match the spatiotemporal scales of satellites, providing an effective method for validating thermal-based ET models and products. Widespread application of this method will promote the optimization of ET retrieval algorithms and improve the accuracy of related products.关键词：Two-wavelength scintillometry;remote sensing evapotranspiration model;Heihe river basin;sensible and latent heat fluxes;temporal-scale matching79|76|0更新时间：2025-12-11 - “本报道聚焦于遥感影像领域，专家基于“高分一号”影像构建高质量多时相湖泊提取数据集，为动态水体监测提供数据支持。”
摘要：Objective This study aims to overcome the limitations of existing public water body datasets, such as single temporal resolution and low annotation accuracy. The objective is to construct a high-quality, multi-temporal lake extraction dataset based on the high-resolution wide-field multispectral imagery from the "GF-1" satellite, which offers improved temporal and spatial coverage of water bodies.Method To achieve this, three study areas with varying levels of dynamic change were selected: Poyang Lake (high dynamic change), Namtso Lake (moderate dynamic change), and Yangcheng Lake (low dynamic change). These areas were covered in four seasons of 2022. The GF-1 wide-field multispectral imagery underwent preprocessing, including radiometric correction, orthorectification, and quick atmospheric correction. For the annotation process, a hybrid strategy combining automated methods with manual visual interpretation was employed to ensure high annotation accuracy.Result The resulting dataset is characterized by multi-temporal data and high annotation accuracy, offering a significant improvement over existing datasets. The overall accuracy of the dataset for all three study areas and across all four seasons exceeded 94%.It provides reliable data for dynamic water body mapping and change monitoring across seasonal variations. Additionally, various water body extraction methods, including threshold segmentation, traditional machine learning algorithms, and deep learning techniques, were employed to validate the dataset’s practical utility. The results demonstrated that the dataset supports the effective training and evaluation of these methods.Conclusion The findings indicate that the constructed multi-temporal lake extraction dataset is highly reliable and can effectively support various water body extraction methods. It provides a robust data foundation for enhancing the performance of dynamic water body extraction algorithms, and contributes valuable data for research in dynamic water body monitoring and mapping using high-resolution remote sensing imagery.关键词：Gaofen-1;Dynamic Water Body;water body extraction;dataset;feature extraction161|166|0更新时间：2025-12-11 - “最新研究揭示了InSAR技术在城市地面沉降监测领域的发展历程，为城市地质安全问题提供解决方案。”
摘要：Objective Land subsidence is one of significant geohazards for urban development, and continuous monitoring and research are essential. The application of Interferometric Synthetic Aperture Radar (InSAR) to urban land subsidence monitoring has advanced markedly across data acquisition, methodologies, accuracy validation, and various applications.Method Building on more than two decades of the authors’ time-series InSAR work in Shanghai, from exploratory application to engineering practice, this paper systematically reviews the development trajectory of urban subsidence monitoring, covering (i) the initial experimental verification stage, (ii) a method development stage based on limited datasets, (iii) infrastructure monitoring enabled by high-resolution SAR data, and (iv) the subsequent expansion toward multi-source, multi-scale InSAR monitoring. Based on the research conducted in Shanghai, the group has extended the study to monitor land subsidence in other regions, such as the Beijing Plain.Result At present, urban land subsidence in Shanghai remains minor and manageable, with annual subsidence kept within 6 mm, which places rigorous demands on the detection of small magnitude deformation.Conclusion Looking ahead, several research directions are expected to further advance InSAR’s development and application in urban settings, including high-precision deformation monitoring in dense high-rise buildings, improved understanding and modeling of radar scattering mechanisms in complex urban environments, scientific interpretation of deformation under multi-factor coupling, and AI-supported information mining and early-warning frameworks.关键词：urban land subsidence;SAR;InSAR;time series InSAR;small dataset method;high-resolution SAR;infrastructure monitoring;multi-scale and multi-sensor monitoring318|934|0更新时间：2025-12-02 - “最新研究揭示中国脱贫县生态与社会经济耦合协调关系，为巩固脱贫成果提供策略。”
摘要：Objective Quantitative analysis of the coupling coordination relationship between the ecological environment and socioeconomic development in China's poverty-alleviated counties, along with the identification of influencing factors, is of great significance for summarizing poverty alleviation experiences and consolidating the achievements of poverty eradication efforts.Method Based on long-term and high-resolution remote sensing datasets, this study constructed a comprehensive evaluation index system for both the ecological environment and socioeconomic development across 832 poverty-alleviated counties in China. It quantitatively assessed the development levels and coupling coordination status of these two dimensions in 2010, 2015, and 2020. Furthermore, a Boosted Regression Tree model was employed to identify the contribution rates of various indicators to the regional coupling coordination development.Result The results indicated that, overall, the ecological environment in poverty-alleviated counties exhibited a higher level of comprehensive development compared with socioeconomic factors. However, the socioeconomic development progressed at a faster pace than ecological improvements. In addition, the average annual growth rates of both dimensions from 2015 to 2020 were higher than those from 2010 to 2015. Spatially, the coupling coordination degree was the highest in northeastern counties and the lowest in the northwest, showing a distribution pattern of “high in the east and low in the west” and a trend of progressive improvement from coastal to inland areas. Most counties were categorized as “economically lagging”. Among all indicators, population size, gross domestic product, and nighttime light intensity made particularly significant contributions to the coupling coordination.Conclusion Drawing from the poverty alleviation paths and practical experiences of typical regions, the study concludes that implementing context-specific industrial poverty alleviation strategies is crucial for accelerating socioeconomic development in poverty-alleviated counties. Establishing diversified, locally distinctive industries is identified as a key approach for consolidating poverty alleviation achievements and preventing a return to poverty.关键词：poverty-alleviated counties;ecological environment;socioeconomic development;Remote sensing dataset;coupling coordination;boosted regression tree model;contribution rate;spatial distribution pattern;industrial poverty alleviation156|128|0更新时间：2025-11-04 -
摘要：1. ObjectiveThe Zoige Wetland covers the largest alpine permafrost peatland in Eurasia, which has an important ecological carbon sink function. Both the freeze-thaw process of permafrost and the carbon cycle of peatlands can cause surface deformation. Monitoring and analyzing deformation can provide important evidence for studying the freeze-thaw and carbon cycle processes. However, current research on surface deformation in this area is relatively scarce. This paper for the first time takes the Zoige peatland as the study area, and the small baseline subset (SBAS) InSAR was used to monitor its deformation. The spatiotemporal characteristics, evolution trends and driving factors of the deformation were comprehensively studied. The health status of peatlands was evaluated, taking into account the deformation distribution and evolution trend.2. MethodThis study obtained 89 ascending and 83 descending Sentinel-1 SAR images from January 2020 to December 2022. The radar line-of-sight (LOS) deformation in the peatland area was extracted by SBAS-InSAR and was verified by comparing with deformation results from the adjacent orbits. The vertical and east-west deformation were obtained by LOS deformation decomposition. The vertical linear cumulative deformation and the seasonal amplitude were extracted by deformation component modeling. At the same time, the historical deformation trend was obtained using the Mann-Kendall trend test and Theil-Sen estimation methods, and the future deformation trend was estimated based on the Hurst index. The spatial and temporal characteristics, change trends of the deformations and health status of the peatland were explored in depth by combining diverse information such as land cover type, surface temperature, and precipitation.3. ResultThe correlation coefficient of the LOS velocity of the overlapping areas between adjacent orbits reaches 0.74, and the root mean square error is ±0.55 mm/a. The vertical and east-west velocity in the study area range from -45 mm/a to 45 mm/a and from -25 mm/a to 25 mm/a, respectively. The vertical deformation is mainly distributed in the peat area being concentrated around the wetlands and water bodies, and the western high-altitude area. The non-peat area in the northwest has relatively obvious eastward deformation affected by elevation and aspect. Seasonal deformation is mostly concentrated in the peat areas, especially around the Cuorewajian Lake and the Manrima Township, with the maximum amplitude of 16.9 mm. The trend test results demonstrate that the areas with significant uplift and subsidence trends account for 51.95% and 26.10% of the total area respectively, while the remaining areas with uplift and subsidence trends account for 8.38% and 5.88% respectively. 75.72% of the area may show anti-continuity trend in the future, and 24.28% of the area may maintain the current trend.4. ConclusionThe deformation of the Zoige peatland exhibits complex characteristics and distribution patterns of subsidence and uplift, linear accumulation and seasonal changes, as well as vertical and horizontal components coexisting. This is mainly related to factors such as freeze-thaw, carbon cycle, land surface temperature, and precipitation. Moreover, the deformation trends of different land types vary significantly, mainly due to the differences in driving factors. Overall, the uplift area of the Zoige peatland is larger than the subsidence area, indicating a good carbon sink function. However, local significant subsidence phenomena occur in areas such as the Cuorewajian Lake surroundings, and are accompanied by a high Hurst index, indicating a significant subsidence persistence. Thus, the peatland in these areas may face up to degradation risks. This study first reveals the complex deformation characteristics, change trends, and influencing factors of the alpine permafrost peatland in Zoige, providing scientific reference for the assessment of ecological functions and vulnerability in this region. It verifies the effectiveness of the SBAS-InSAR technology in monitoring the surface deformation of large-scale permafrost peatlands.关键词：SBAS-InSAR;Freeze-thaw Peatland;deformation monitoring;Spatiotemporal Deformation Characteristics;Seasonal Variation;Driving Factors of Deformation;Assessment of Peatland Health Status152|728|0更新时间：2025-10-28 -
摘要：Objective Semantic segmentation of remote sensing images (RSIs) plays a crucial role in land cover and land use classification, urban planning, and change detection. As a highly promising unsupervised learning method, domain adaptation has significantly accelerated the advancement of RSI segmentation. However, current models often rely on the limited feature learning capability of single-task approaches, making it difficult to accurately distinguish hard-to-classify regions in RSIs. To address this issue, a multi-task learning domain adaptive network (MTLDANet) is proposed, which jointly learns semantic and elevation information in RSIs, improving segmentation performance.Method The method feeds task-specific semantic and elevation features into a cross-task feature correlation learning module to explore latent correlations between tasks, thereby enhancing task-specific feature representations. A hybrid consistency learning module, guided by pseudo-labels, is employed to improve pseudo-label quality and achieve global domain alignment. Additionally, entropy-guided category-level alignment enhances the separability of challenging categories.Result The proposed method is evaluated on four cross-scene RSI segmentation experiments using the ISPRS 2D and US3D datasets.Conclusion Results show that the method outperforms existing domain adaptation approaches, demonstrating significant advantages in various complex cross-domain scenarios.关键词：semantic segmentation;unsupervised domain adaptation;remote sensing imagery;Multi-task learning;elevation information;semantic information;pseudo-label;Entropy291|584|0更新时间：2025-10-28 -
摘要：Objective The objective of this study is to address the limitations of existing datasets used for building height estimation from optical and SAR remote sensing imagery. Current datasets often suffer from small sample sizes, limited geographic diversity, and a lack of openness, making them insufficient for supporting deep learning-based remote sensing applications—especially for large-scale studies in China. Accurately estimating building heights is critical for understanding urban morphology and optimizing urban stock space, thus necessitating the development of a more comprehensive, representative, and accessible dataset. Methods To overcome these issues, this paper constructs a new dataset named BHDSI, specifically designed for building height regression tasks. The dataset comprises 5,606 samples from the central urban areas of 62 cities across China, making it the largest building height dataset for the country in terms of geographic coverage. It includes both Sentinel-1 and Sentinel-2 imagery along with true building height values, with each sample having a spatial resolution of 256×256 pixels. This provides a richer spatial context compared to existing datasets with smaller sample sizes, such as 64×64. The dataset encompasses a wide range of scenarios, including urban and rural areas, ensuring better representation of spatial features. Result Experimental evaluations demonstrate that the BHDSI dataset leads to superior performance in building height regression tasks when compared to other similar datasets, across various deep learning networks. The results also indicate that estimation accuracy tends to be higher in regions with lower building heights. Furthermore, the study finds that using a U-Net decoder structure in the network architecture contributes to higher prediction precision, highlighting the importance of decoder design in deep learning-based height estimation. Conclusion The BHDSI dataset significantly advances the field of building height estimation by offering a large-scale, diverse, and high-quality resource tailored for deep learning. Its broad coverage, balanced height distribution, and open accessibility make it better suited for training and evaluating deep neural networks than previously available datasets. The study confirms that both data quality and network architecture, especially decoder design, play vital roles in improving estimation accuracy, and BHDSI serves as a strong foundation for future research in this domain.关键词：Sentinel imagery;building height;dataset;deep learning;convolutional neural network307|1055|0更新时间：2025-10-28 -
摘要：Multi-temporal hyperspectral images have a wide range of applications in change detection due to their rich spectral features and image details. Traditional hyperspectral change detection algorithms based on supervised learning often rely on a large number of labeled samples, which requires a large sample annotation cost. In recent years, although research has explored the problem of change detection under limited labeled samples, there are still many aspects that need further exploration. Existing methods often fail to fully tap into the potential of limited labeled samples, and there are also shortcomings in extracting changing features. Therefore, we have developed a new network architecture aimed at more effectively utilizing limited labeled samples and focusing on extracting differential features to enhance information related to changes.In this paper, we propose a joint central difference feature and spatial-spectral attention network (JCDS2AN) for hyperspectral image change detection, which can alleviate the fluctuation of changing features under sample constraints and learn representative changing features using limited labeled samples. In JCDS2AN, a multi-scale spatial-spectral attention block was designed to capture multi-scale spatial and spectral features, and a differential center pixel exchange strategy guided by differential features was proposed to achieve efficient information exchange between differential features and two temporal features.Experimental results on three publicly available hyperspectral image datasets show that the proposed JCDS2AN outperforms the state-of-the-art methods in hyperspectral change detection. When utilizing only 1% of the training samples, the method achieved optimal Kappa and OA of 95.90% and 98.30%, respectively, on the Farmland dataset. Ablation experiments were conducted for each proposed module to demonstrate their effectiveness. This approach is capable of extracting discriminative deep change semantic information, with both qualitative and quantitative results surpassing those of other advanced networks.关键词：hyperspectral image;remote sensing images;change detection;Multi-scale Features;differential feature guidance;center pixel154|304|0更新时间：2025-10-28 - “SAMSNet在遥感图像道路提取领域取得突破，提升了IoU和F1-score等指标，为智慧城市建设提供技术支撑。”
摘要：Automatic road extraction from high-resolution remote sensing images plays a crucial role in applications such as smart cities, intelligent transportation, and autonomous driving. However, existing methods often suffer from issues like fragmentation and poor connectivity in the extracted road networks, especially under complex scenarios with occlusions, shadows, and large scale variations. This study aims to develop a robust deep learning model capable of extracting continuous and complete road networks from high-resolution remote sensing imagery by effectively integrating multi-scale contextual information and attention mechanisms.An improved encoder-decoder network named SAMSNet is proposed.. The encoder is based on ResNeSt-50, which utilizes a split-attention mechanism to enhance cross-channel feature interaction and capture richer semantic representations. To expand the receptive field and aggregate multi-scale context without losing spatial details, a cascaded parallel dilated convolution block (Dblock) is introduced in the central part of the network. Furthermore, a Multi-Scale Channel Attention Module (MS-CAM) is incorporated into the skip connections to simultaneously emphasize both global and local road features, improving the model's ability to handle extreme scale variations. The network is trained using a combined loss function of binary cross-entropy and Dice loss to address class imbalance and emphasize boundary accuracy.Extensive experiments were conducted on three public road extraction datasets DeepGlobe, Massachusetts, and GRSet. SAMSNet achieved state-of-the-art performance across all datasets. On the DeepGlobe dataset, it attained an IoU of 74.48% and an F1-score of 85.37%, significantly outperforming other models such as U-Net, D-LinkNet, and transformer-based approaches. Similar improvements were observed on the Massachusetts dataset, with IoU and F1-score reaching 66.61% and 79.96%, respectively. Transfer learning experiments on the GRSet dataset further demonstrated the strong generalization capability of SAMSNet, where it achieved the highest IoU (55.55%) and F1-score (60.71%) among all compared models. Ablation studies confirmed the individual contributions of the Dblock and MS-CAM modules to the overall performance.关键词：remote sensing images;road extraction;semantic segmentation;ResNeSt-50;Dispersed Attention;Multi-Scale Channel Attention;dilated convolution124|183|0更新时间：2025-10-27 - “最新研究整合InSAR技术，识别青海省化隆县334个滑坡隐患，为地质灾害防治提供技术支持。”
摘要：Objectives Landslide detection and deformation monitoring are critical for geological hazard prevention and risk mitigation. This study proposes an integrated framework for landslide detection and three-dimensional (3D) deformation monitoring using Interferometric Synthetic Aperture (InSAR). The proposed framework comprises three key components: (1) Time-series InSAR data processing, (a) landslide detection based on InSAR results, optical images and C-index, (3) 3D deformation monitoring. We apply the proposed framework to Hualong County in Qinghai province, a landslide prone area.Methods Initially, we employ Multi-temporal InSAR (MT-InSAR) to analyze both the ascending and descending Sentinel-1 satellite images acquired between Jan. 2021 and Jun. 2023, encompassing Hualong County in Qinghai Province. Next, deformation rate maps are generated and cross-validated with field measurements from Global Navigation Satellite System (GNSS) station. Landslide identification is performed through the integration of InSAR-derived deformation signals, high-resolution Google Earth™ imagery, and C-index of each potential deformation areas. Subsequently, we apply the Aspect Parallel Flow Model (APFM) to calculate the 3D displacement field of representative landslides.Results The deformation results derived from both InSAR and on-site equipment exhibit strong agreement, and the standard deviation of the obtained average deformation rates is 5 mm/a for both ascending and descending images, showing high reliability of the obtained InSAR deformation results. Through the integration of InSAR deformation rate maps and Google EarthTM imagery, we detected a total of 334 landslides. Among these, 233 landslides were discernible using ascending data, 265 with descending data, and 164 were detectable with both ascending and descending datasets. The total area of the detected landslides is about 95.56 km2. The slope gradients of the detected landslides range between 5° and 40°, with 184 landslides posing direct threats to infrastructure (e.g., buildings, roads) and natural features (e.g., rivers). There is a notably lower count of landslides detected in the near north-south direction compared to other orientations, suggesting that InSAR might exhibit reduced sensitivity to deformations associated with landslides occurring along this axis. Theoretically, it has been demonstrated that observational errors can notably influence the 3D displacement field obtained from APFM, particularly in the context of landslides occurring in a nearly north-south direction. Utilizing APFM, the 3D deformation time series of the Anjuhu landslide in Chuma Township, which encompasses the largest area, was calculated. Analysis reveals that the horizontal deformation significantly outweighed the vertical deformation, with the maximum cumulative horizontal displacement surpassing 1 meter. The landslide presents a threat to two villages, a provincial road, and agricultural areas, necessitating ongoing monitoring.Conclusions This study demonstrates the effectiveness of InSAR for regional-scale landslide detection and 3D deformation monitoring while also highlighting its limitations, particularly in areas with unfavorable slope orientations. The results illustrate both the benefits and limitations of InSAR in landslide monitoring, offering practical examples that can guide county-level efforts in landslide identification and 3D deformation monitoring, while also providing technical support.关键词：InSAR;SBAS;landslide detection;C-index;landslide monitoring;three-dimensional deformation;Hualong County;ascending and descending214|251|0更新时间：2025-09-30 - “最新研究突破东北黑土区侵蚀沟监测难题，提出循环自训练框架，显著提升跨时相提取性能，为土地保护管理提供新方案。”
摘要：Soil erosion poses a serious threat to the black soil area of Northeast China, and gully erosion is one of its main manifestations. At present, remote sensing technology has been widely used in the monitoring and management of gully erosion, and a large number of labeled historical survey data have been accumulated. However, how to use these historical data to reliably extract gully information from the latest data captured by different sensors at different times is still an urgent technical problem to be solved. Therefore, the objective of this paper is to develop an effective method to achieve reliable cross-temporal gully extraction and provide technical support for black soil area land protection and management.Based on the above objective, this paper proposes a cyclic self - training framework (CSTF). It employs an iterative self-training approach to achieve reliable cross-temporal gully extraction. In each self-training iteration, an object-level pseudo-label generation strategy is designed to ensure high-quality pseudo-labels for the latest data. Additionally, a loss function based on pseudo-label confidence factors is introduced to effectively mitigate the adverse effects of pseudo-label noise. These two elements complement each other and significantly enhance the model’s performance in the task of cross-temporal gully erosion extraction.In the experimental section, Huachuan County, Heilongjiang Province, China, was selected as the study area, and the following results were drawn: (1) The characteristic differences between historical and current data present significant challenges for cross-temporal extraction tasks. Compared to traditional fully supervised methods, Unsupervised Domain Adaptation (UDA) methods offer superior performance. Additionally, the self-training methods demonstrate greater robustness than invariant representation learning, thus justifying its use in cross-temporal extraction studies. (2) For self-training methods, the quality of pseudo-labels is a critical factor influencing performance. To address this, a series of improvement strategies are proposed, leading to the best results in both accuracy assessments and visual interpretation. Specifically, the Intersection over Union (IoU) is 7.39% and 7.90% higher than the second-best methods in Experiments 1 and 2, respectively. Furthermore, these strategies are shown to be effective, necessary, and compatible, as demonstrated through detailed ablation experiments. Regarding the analysis of algorithm complexity and operational efficiency, the proposed CSTF can not only ensure accurate extraction results but also offer high efficiency, meeting the actual monitoring requirements for gully erosion.In conclusion, the proposed CSTF provides robust technical support for cultivated land conservation in black soil regions and offers a promising approach for sustainable land management. Currently, the CSTF only deals with the binary classification of erosion gullies and non - eroded areas. Future research will expand the framework to recognize gullies at different developmental stages, facilitating more refined monitoring and analysis of erosion gullies.关键词：Soil erosion;black soil area of Northeast China;gully erosion;cross-temporal extraction;self-training;object-level pseudo-label generation strategy;pseudo-label credibility factor;pseudo-label noise;land protection and management67|137|0更新时间：2025-09-30 -
Research on Cascade Learning for Object Detection in Remote Sensing Images under Foggy Conditions AI导读
“在遥感图像领域，专家提出了级联学习的雾下目标检测方法，建立了轻量化去雾子网络，实现去雾与目标检测的协同优化，为解决雾天条件下目标检测问题提供解决方案。”
摘要：Under foggy conditions, atmospheric scattering reduces the illumination intensity in images, which leads to a decrease in the contrast of remote sensing images and affects the performance of object detection models. Existing research has addressed this issue through two strategies: training models on foggy data or using image dehazing as a preprocessing step. However, the dehazing process can result in loss of feature, and it is difficult to ensure a consistently positive correlation between dehazing results and object detection tasks, i.e., that the dehazing results are beneficial for object detection. To address this issue, a Cascade Learning Foggy Object Detection Method (CL-FODM) is proposed in this paper. The method establishes a lightweight dehazing sub-network combining CNN and Transformer, which can obtain clearer dehazed features and provide more salient semantic information for the object detection task. A multi-task loss function guided by feature perception is constructed to more precisely mine discriminative target semantic features at the feature level, achieving collaborative optimization between dehazing and object detection and solving the semantic inconsistency between low-level and high-level tasks. Experimental results show that the CL-FODM proposed in this paper outperforms both the original model and the cascaded model in terms of evaluation metrics and visual detection effects.关键词：remote sensing imagery;object detection;Dehazing Model;deep learning;Cascade Learning216|164|0更新时间：2025-09-08
- Postal code：100190
- Tel：010-58887052 Email：nrsb@aircas.ac.cn
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备20021838号-9
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0