Search Results (11,609)

Lakes on the Qinghai–Tibet Plateau are important indicators of global climate change, and variations in their water storage strongly influence regional hydrological cycles and ecosystems. However, existing studies have largely focused on relative changes in lake volume, while the precise quantification of absolute water storage remains insufficient, largely due to the lack of long-term, high-accuracy water storage time series. Constrained by harsh natural conditions and limited in situ observations, conventional approaches struggle to achieve the accurate long-term monitoring of lake water storage across the Plateau. To address this challenge, we propose a DEM-based underwater topography extrapolation method. Under the assumption of continuity between surrounding onshore terrain and submerged lakebed morphology, nearshore DEM data are extrapolated to reconstruct lake bathymetry. By integrating multi-source remote sensing observations of lake area and water level, we estimate and reconstruct 30-year absolute water storage time series for 120 Plateau lakes larger than 50 km². This method does not require measured water depth data and is particularly suitable for data-scarce, topographically complex, high-altitude lake regions, effectively overcoming key limitations of conventional methods used for absolute water storage monitoring. Validation shows strong agreement between our estimates and an independent validation dataset, with an overall correlation coefficient of 0.95; the reconstructed time series are highly reliable, with correlation coefficients exceeding 0.6. During the study period, the total lake water storage of the Qinghai–Tibet Plateau exhibited a significant increasing trend, with a cumulative growth of approximately 137.297 billion m³, representing a 20.73% increase, and showing notable spatial heterogeneity. The water storage dataset constructed in this study provides reliable data support for research on water cycles, climate change assessment, and regional water resource management on the Qinghai–Tibet Plateau. Full article

The simultaneous measurement of strain and temperature using Fiber Bragg Grating (FBG) sensors presents a significant challenge due to the intrinsic cross-sensitivity of the Bragg wavelength. While recent studies have increasingly employed “black-box” machine learning algorithms to address this ambiguity, such approaches often overlook the physical limitations of the sensor’s spectral response. This paper challenges the assumption that advanced algorithms alone can compensate for data that is physically ambiguous. We propose a “Sensor-Algorithm Co-Design” methodology, demonstrating that robust discrimination is achievable only when the sensor architecture exhibits a unique, orthogonal physical signature. Using a rigorous Transfer Matrix Method (TMM) and 4 × 4 polarization analysis, we evaluate three distinct architectures. Quantitative analysis reveals that a standard Quadratically Chirped FBG (QC-FBG) functions as an “ill-conditioned baseline” failing to distinguish measurands due to feature space collapse (K_{cond >} 4600). Conversely, we validate two robust co-designs: (1) An Amplitude-Modulated Superstructure FBG (S-FBG) paired with an Artificial Neural Network (ANN), utilizing thermally induced duty-cycle variations to achieve high accuracy (~3.4 °C error) under noise; and (2) A Polarization-Diverse Inverse-Gaussian FBG (IG-FBG) paired with a 4 × 4 K-matrix, exploiting strain-induced birefringence (K_cond ≈ 64). Furthermore, we address the data scarcity issue in AI-driven sensing by introducing a Physics-Informed Neural Network (PINN) strategy. By embedding TMM physics directly into the loss function, the PINN improves data efficiency by 2.2× compared to standard models, effectively bridging the gap between physical modeling and data-driven inference, addressing the critical data scarcity bottleneck identified in recent optical sensing roadmaps. Full article

Farmland shelterbelts are crucial elements in safeguarding agricultural ecological security and sustainable development, with their precise extraction being vital for regional ecological monitoring and precision agriculture management. However, constrained by their narrow linear distribution, complex farmland backgrounds, and spectral confusion issues, traditional remote sensing methods encounter significant challenges in terms of accuracy and generalization capability. In this study, six representative deep learning semantic segmentation models—U-Net, Attention U-Net (AttU_Net), ResU-Net, U²-Net, SwinUNet, and TransUNet—were systematically evaluated for farmland shelterbelt extraction using high-resolution Gaofen-6 imagery. Model performance was assessed through four-fold cross-validation and independent test set validation. The results indicate that convolutional neural network (CNN)-based models show overall better performance than Transformer-based architectures; on the independent test set, the best-performing CNN model (U-Net) achieved a Dice Similarity Coefficient (DSC) of 91.45%, while the lowest DSC (88.86%) was obtained by the Transformer-based TransUNet model. Among the evaluated models, U-Net demonstrated a favorable balance between accuracy, stability, and computational efficiency. The trained U-Net was applied to large-scale farmland shelterbelt mapping in the study area (Alar City, Xinjiang), achieving a belt-level visual accuracy of 95.58% based on 385 manually interpreted samples. Qualitative demonstrations in Aksu City and Shaya County illustrated model transferability. This study provides empirical guidance for model selection in high-resolution agricultural remote sensing and offers a feasible technical solution for large-scale and precise farmland shelterbelt extraction. Full article

Urban villages (UVs) are widespread in rapidly urbanizing regions, but their fine-grained delineation from high-resolution remote sensing imagery remains a challenge due to complex spatial textures and ambiguous boundaries. To address this issue, this paper proposes TransUV, a TransNeXt-based encoder–decoder segmentation framework tailored to UV extraction. At the encoder front end, a Multi-level Feature Enhancement Module (MFEM) injects boundary- and texture-aware inductive bias by combining Laplacian-of-Gaussian (LoG) filtering with Gaussian smoothing, which strengthens edge responses while suppressing noise. At the decoder stage, we design a lightweight SegUV decoder equipped with an Advanced Attention Fusion Module (AAFM) that adaptively fuses multi-scale features using complementary channel, spatial, and directional attention. Experiments on 0.5 m imagery from two Chinese cities demonstrate that TransUV achieves an mIoU of 86.67% and an overall accuracy of 92.98%, significantly outperforming other mainstream models. Full article

Radar systems with exceptional anti-jamming performance are critical to meeting the high-performance requirements of future intelligent sensing systems. To address the deception jamming challenges encountered by intelligent sensing systems environments, a multi-parameter agile waveform is designed. The proposed waveform exhibits high flexibility across three dimensions—pulse width, pulse repetition interval, and carrier frequency. Compared to traditional single-parameter or two-parameter agile waveforms, which vary only one or two parameters, this multi-parameter approach significantly enhances anti-jamming performance by disrupting periodicity and providing higher flexibility in dynamic interference environments. To address the complex signal characteristics induced by multi-parameter agility, we further develop a low-complexity signal processing method based on a segmented multiple signal classification (MUSIC) algorithm, which accurately extracts Doppler information from pulse-compressed slow-time data to achieve high-precision velocity estimation. Both theoretical derivations and simulation results demonstrate that, compared with the conventional compressed sensing orthogonal matching pursuit method and the conventional MUSIC method that operate on the entire signal, our segmented approach divides the signal into smaller segments, reducing computational complexity and improving velocity estimation accuracy. Notably, even in high-intensity, densely jammed environments, the system reliably extracts target information. Full article

Wildfires, which encompass all fires that occur outside urban areas, represent one of the most frequent forms of natural disaster worldwide. This study presents the wildfire occurrence across the territory of Southeastern Europe, covering an area of 800,000 km² (Greece, Romania, Serbia, Slovenia, Croatia, Bosnia and Herzegovina, Montenegro, Albania, North Macedonia, Bulgaria, and Moldova). The research applies geospatial artificial intelligence techniques, based on the integration of machine learning (Random Forest (RF), XGBoost), deep learning (Deep Neural Network (DNN), Kolmogorov–Arnold Networks (KAN)), remote sensing (Sentinel-2, VIIRS), and Geographic Information Systems (GIS). From the geospatial database, 11 natural and anthropogenic criteria were analyzed, along with a wildfire inventory comprising 28,952 historical fire events. The results revealed that areas of very high susceptibility were most prevalent in Greece (10.5%), while the smallest susceptibility percentage was recorded in Slovenia (0.2%). Among the applied models, RF demonstrated the highest predictive performance (AUC = 90.7%), whereas XGBoost, DNN, and KAN achieved AUC values ranging from 86.7% to 90.5%. Through a SHAP analysis, it was determined that the most influential factors were global horizontal irradiation, elevation, and distance from settlements. The obtained results hold international significance for the implementation of preventive wildfire protection measures. Full article

With Transformers achieving outstanding performance on individual remote sensing (RS) tasks, we are now approaching the realization of a unified model that excels across multiple tasks through multi-task learning (MTL). Compared to single-task approaches, MTL methods offer improved generalization, enhanced scalability, and greater practical applicability. Recently, vision-language models (VLMs) have achieved promising results in RS image understanding, grounding, and ultra-high-resolution (UHR) image reasoning, respectively. Moreover, the unified text-based interface demonstrates significant potential for MTL. Hence, in this work, we present RSCoVLM, a simple yet flexible VLM baseline for RS MTL. Firstly, we create the data curation procedure, including data acquisition, offline processing and integrating, as well as online loading and weighting. This data procedure effectively addresses complex RS data enviroments and generates flexible vision-language conversations. Furthermore, we propose a unified dynamic-resolution strategy to address the diverse image scales inherent in RS imagery. For UHR images, we introduce the Zoom-in Chain mechanism together with its corresponding dataset, LRS-VQA-Zoom. The strategies are flexible and effectively mitigate the computational burdens. Additionally, we significantly enhance the model’s object detection capability and propose a novel evaluation protocol that ensures fair comparison between VLMs and conventional detection models. Extensive experiments demonstrate that RSCoVLM achieves state-of-the-art performance across diverse tasks, outperforming existing RS VLMs and even rivaling specialized expert models. All the training and evaluating tools, model weights, and datasets have been fully open-sourced to support reproducibility. We expect that this baseline will promote further progress toward general-purpose RS models. Full article

Remote sensing provides an efficient approach for monitoring ecosystem dynamics in the arid and semi-arid regions of Xinjiang, yet traditional forest-land extraction methods (e.g., spectral indices, threshold segmentation) show limited adaptability in complex environments affected by terrain shadows, cloud contamination, and spectral confusion with grassland or cropland. To overcome these limitations, this study used three convolutional neural network-based models (FCN, DeepLabV3+, and PSPNet) for accurate forest-land extraction. Four tri-band training datasets were constructed from Sentinel-2 imagery using combinations of visible, red-edge, near-infrared, and shortwave infrared bands. Results show that the FCN model trained with B4–B8–B12 achieves the best performance, with an mIoU of 89.45% and an mFscore of 94.23%. To further assess generalisation in arid landscapes, ESA WorldCover and Dynamic World products were introduced as benchmarks. Comparative analyses of spatial patterns and quantitative metrics demonstrate that the FCN model exhibits robustness and scalability across large areas, confirming its effectiveness for forest-land extraction in arid regions. This study innovatively combines band combination optimization strategies with multiple deep learning models, offering a novel approach to resolving spectral confusion between forest areas and similar vegetation types in heterogeneous arid ecosystems. Its practical significance lies in providing a robust data foundation and methodological support for forest monitoring, ecological restoration, and sustainable land management in Xinjiang and similar regions. Full article

To achieve precise in-hand manipulation and feedback control using soft robotic fingers, it is essential to accurately measure their deformable structures. In particular, estimating the three-dimensional shape of a soft finger under contact conditions is a critical challenge, as the deformation state directly affects manipulation reliability. However, nonlinear deformations and occlusions arising from interactions with external objects make the estimation difficult. To address these issues, we propose a soft finger structure that integrates small magnets and magnetic sensors inside the body, enabling the acquisition of rich deformation information in both contact and non-contact states. The design provides a 15-dimensional time-series signal composed of motor angles, motor currents, and magnetic sensor outputs as inputs for shape estimation. Built on the sensing signals, we propose a mode-selection-based learning approach that outputs multiple candidate shapes and selects the correct one. The proposed network predicts the three-dimensional positions of four external markers attached to the finger, which serve as a proxy representation of the finger’s shape. The network is trained in a supervised manner using ground-truth marker positions measured by a motion capture system. The experimental results under both contact and non-contact conditions demonstrate that the proposed method achieves an average estimation error of approximately 4 mm, outperforming conventional one-shot regression models that output coordinates directly. The integration of magnetic sensing is demonstrated to be able to enable accurate recognition of contact states and significantly improve stability in shape estimation. Full article

Urban surface subsidence is primarily induced by intensive above-ground and underground construction activities and excessive groundwater extraction. Integrating InSAR techniques for safety monitoring of urban subway infrastructure is therefore of great significance for urban safety and sustainable development. However, single-track high-spatial-resolution SAR imagery is insufficient to achieve full coverage over large urban areas, and direct mosaicking of inter-track InSAR results may introduce systematic biases, thereby compromising the continuity and consistency of deformation fields at the regional scale. To address this issue, this study proposes an inter-track InSAR correction and mosaicking approach based on the mean vertical deformation difference within overlapping areas, aiming to mitigate the overall offset between deformation results derived from different tracks and to construct a spatially continuous urban surface deformation field. Based on the fused deformation results, subsidence characteristics along subway lines and in key urban infrastructures were further analyzed. The main urban area and the eastern and western new districts of Zhengzhou, a national central city in China, were selected as the study area. A total of 16 Radarsat-2 SAR scenes acquired from two tracks during 2022–2024, with a spatial resolution of 3 m, were processed using the SBAS-InSAR technique to retrieve surface deformation. The results indicate that the mean deformation rate difference in the overlapping areas between the two SAR tracks is approximately −5.54 mm/a. After applying the difference-constrained correction, the coefficient of determination (R²) between the mosaicked InSAR results and leveling observations increased to 0.739, while the MAE and RMSE decreased to 4.706 and 5.538 mm, respectively, demonstrating good stability in achieving inter-track consistency and continuous regional deformation representation. Analysis of the corrected InSAR results reveals that, during 2022–2024, areas exhibiting uplift and subsidence trends accounted for 37.6% and 62.4% of the study area, respectively, while the proportions of cumulative subsidence and uplift areas were 66.45% and 33.55%. In the main urban area, surface deformation rates are generally stable and predominantly within ±5 mm/a, whereas subsidence rates in the eastern new district are significantly higher than those in the main urban area and the western new district. Along subway lines, deformation rates are mainly within ±5 mm/a, with relatively larger deformation observed only in localized sections of the eastern segment of Line 1. Further analysis of typical zones along the subway corridors shows that densely built areas in the western part of the main urban area remain relatively stable, while building-concentrated areas in the eastern region exhibit a persistent relative subsidence trend. Overall, the results demonstrate that the proposed inter-track InSAR mosaicking method based on the mean deformation difference in overlapping areas can effectively support subsidence monitoring and spatial pattern identification along urban subway lines and key regions under relative calibration conditions, providing reliable remote sensing information for refined urban management and infrastructure risk assessment. Full article

Small-object detection in UAV remote sensing faces common challenges such as tiny target size, blurred features, and severe background interference. Furthermore, single imaging modalities exhibit limited representation capability in complex environments. To address these issues, this paper proposes CTU-YOLO, a UAV-based small-object detection algorithm built upon cross-modal feature fusion and lightweight upsampling. The algorithm incorporates a dynamic and adaptive cross-modal feature fusion (DCFF) module, which achieves efficient feature alignment and fusion by combining frequency-domain analysis with convolutional operations. Additionally, a lightweight upsampling module (LUS) is introduced, integrating dynamic sampling and depthwise separable convolution to enhance the recovery of fine details for small objects. Experiments on the DroneVehicle and LLVIP datasets demonstrate that CTU-YOLO achieves 73.9% mAP on DroneVehicle and 96.9% AP on LLVIP, outperforming existing mainstream methods. Meanwhile, the model possesses only 4.2 MB parameters and 13.8 GFLOPs computational cost, with inference speeds reaching 129.9 FPS on DroneVehicle and 135.1 FPS on LLVIP. This exhibits an excellent lightweight design and real-time performance while maintaining high accuracy. Ablation studies confirm that both the DCFF and LUS modules contribute significantly to performance gains. Visualization analysis further indicates that the proposed method can accurately preserve the structure of small objects even under nighttime, low-light, and multi-scale background conditions, demonstrating strong robustness. Full article

The existing multimodal image classification methods often suffer from several key limitations: difficulty in effectively balancing local detail and global topological relationships in hyperspectral image (HSI) feature extraction; insufficient multi-scale characterization of terrain features from light detection and ranging (LiDAR) elevation data; and neglect of deep inter-modal interactions in traditional fusion methods, often accompanied by high computational complexity. To address these issues, this paper proposes a comprehensive deep learning framework combining convolutional neural network (CNN), a graph convolutional network (GCN), and wavelet transform for the joint classification of HSI and LiDAR data, including several novel components: a Spectral Graph Mixer Block (SGMB), where a CNN branch captures fine-grained spectral–spatial features by multi-scale convolutions, while a parallel GCN branch models long-range contextual features through an enhanced gated graph network. This dual-path design enables simultaneous extraction of local detail and global topological features from HSI data; a Spatial Coordinate Block (SCB) to enhance spatial awareness and improve the perception of object contours and distribution patterns; a Multi-Scale Elevation Feature Extraction Block (MSFE) for capturing terrain representations across varying scales; and a Bidirectional Frequency Attention Encoder (BiFAE) to enable efficient and deep interaction between multimodal features. These modules are intricately designed to work in concert, forming a cohesive end-to-end framework, which not only achieves a more effective balance between local details and global contexts but also enables deep yet computationally efficient interaction across features, significantly strengthening the discriminability and robustness of the learned representation. To evaluate the proposed method, we conducted experiments on three multimodal remote sensing datasets: Houston2013, Augsburg, and Trento. Quantitative results demonstrate that our framework outperforms state-of-the-art methods, achieving OA values of 98.93%, 88.05%, and 99.59% on the respective datasets. Full article

Aiming at the problem that the current traditional apple yield estimation methods rely on manual investigation and do not make full use of multi-source information, this paper proposes an apple orchard yield prediction framework combining multispectral remote sensing features and ground features. The framework is oriented to the demand of yield prediction at different scales. It can not only realize the prediction of apple yield at the district and county scales, but also modify the prediction results of small-scale orchards based on the acquisition of orchard features. The framework consists of three parts, namely, apple orchard planting area extraction, district and county large-scale yield prediction and small-scale orchard yield prediction correction. (1) During apple orchard planting area extraction, the samples of some apple planting areas in the study area were obtained through field investigation, and the orchard and non-orchard areas were classified and discriminated, providing a spatial basis for the collection of subsequent yield prediction-related data. (2) In the large-scale yield prediction of districts and counties, based on the obtained orchard-planting areas, the corresponding multispectral remote sensing features and environmental features were obtained using Google Earth engine platform. In order to avoid the noise interference caused by local pixel differences, the obtained data were median synthesized, and the feature set was constructed by combining the yield and other information. On this basis, the feature set was divided and sent to Apple Orchard Yield Prediction Network (APYieldNet) for training and testing, and the district and county large-scale yield prediction model was obtained. (3) During the part of small-scale orchard yield prediction correction, the optimal model for large-scale yield prediction at the district and county levels is utilized to forecast the yield of the entire planting area and the internal local sampling areas of the small-scale orchard. Within the local sampling areas, the number of fruits is identified through the YOLO-A model, and the actual yield is estimated based on the empirical single fruit weight as a ground feature, which is used to calculate the correction factor. Finally, the proportional correction method is employed to correct the error in the prediction results of the entire small-scale orchard area, thus obtaining a more accurate yield prediction for the small-scale orchard. The experiment showed that (1) the yield prediction model APYieldNet (MAE = 152.68 kg/mu, RMSE = 203.92 kg/mu) proposed in this paper achieved better results than other methods; (2) the proposed YOLO-A model achieves superior detection performance for apple fruits and flowers in complex orchard environments compared to existing methods; (3) in this paper, through the method of proportional correction, the prediction results of APYieldNet for small-scale orchard are closer to the real yield. Full article

Global water environment monitoring urgently requires remote sensing data with high temporal resolution and wide spatial coverage. However, current space-borne ocean color spectrometers still face a significant trade-off among spatial resolution, swath width, and system compactness, which limits the large-scale deployment of satellite constellations. To address this challenge, this study developed a lightweight high-resolution spectral imager (LHRSI) with a total mass of less than 25 kg and power consumption below 80 W. The visible (VIS) camera adopts an interleaved dual-field-of-view and detectors splicing fusion design, while the shortwave infrared (SWIR) camera employs a transmission-type focal plane with staggered detector arrays. Through the field-of-view (FOV) optical design, the instrument achieves swath widths of 207.33 km for the VIS bands and 187.8 km for the SWIR bands at an orbital altitude of 500 km, while maintaining spatial resolutions of 12 m and 24 m, respectively. On-orbit imaging results demonstrate that the spectrometer achieves excellent performance in both spatial resolution and swath width. In addition, preliminary analysis using index-based indicators illustrates LHRSI’s potential for observing chlorophyll-related features in water bodies. This research not only provides a high-performance, miniaturized spectrometer solution but also lays an engineering foundation for developing low-cost, high-revisit global ocean and water environment monitoring constellations. Full article

Edge detection is a key building block in real-time image-processing applications such as drone-based infrastructure inspection, autonomous navigation, and remote sensing. However, its computational cost remains a challenge for resource-constrained embedded systems. This work presents a hardware-accelerated edge detection architecture based on a homogeneous 2D-mesh Network-on-Chip (NoC) integrating systolic arrays to efficiently perform the convolution operations required by the Sobel filter. The proposed architecture was first developed and validated as a 3 × 3 mesh prototype on FPGA (Xilinx Zynq-7000, Zynq-7010, XC7Z010-CLG400A, Zybo board, utilizing 26,112 LUTs, 24,851 flip-flops, and 162 DSP blocks), achieving a throughput of 8.8 Gb/s with a power consumption of 0.79 W at 100 MHz. Building upon this validated prototype, a reduced 2 × 2 node cluster with 14-bit word width was subsequently synthesized at the physical level as a proof-of-concept using the OpenLane RTL-to-GDSII open-source flow targeting the SkyWater 130 nm PDK (sky130A). Post-layout analysis confirms the manufacturability of the design, with a total power consumption of 378 mW and compliance with timing constraints, demonstrating the feasibility of mapping the proposed architecture to silicon and its suitability for drone-based infrastructure monitoring applications. Full article