Search Results (2,331)

Accurate and up-to-date mapping of built-up areas is of great importance for sustainable urban planning, disaster management, and the monitoring of environmental changes. In this study, a residual U-Net-based deep learning architecture named FiveBandTTA is proposed for built-up area segmentation from Sentinel-2 multispectral satellite imagery. The proposed model aims to simultaneously learn spatial and spectral features by jointly processing RGB, NIR (B8), and SWIR (B11) bands within the same encoder–decoder structure. The model incorporates standard residual blocks following the conventional residual learning principle, multi-level skip connection mechanisms, and TTA-based inference strategies. Within the scope of the study, a multi-temporal built-up area dataset was constructed from Sentinel-2 imagery acquired over Kocaeli Province. The performance of the proposed model was comparatively evaluated against RGB Baseline, FiveBand Single, DeepLabV3+, and SegFormer models. Experimental results demonstrated that the proposed model achieved the highest segmentation performance among all compared approaches, obtaining 0.8447 IoU, 0.9124 Dice, and 0.9249 Precision scores. It was observed that the use of multispectral bands together with the residual encoder–decoder structure may contribute to improved representation of small-scale built-up regions and complex boundary structures. Furthermore, the comparative experiments indicated that the NIR and SWIR bands provide complementary spectral information for distinguishing built-up areas, while the TTA-based inference strategy may contribute to improved segmentation stability and prediction consistency. Overall, the obtained results demonstrate that the proposed approach is an effective and robust method for built-up area segmentation from medium-resolution Sentinel-2 imagery. Full article

Understanding how brain activity varies across neurological and neurodevelopmental disorders requires tools capable of revealing patterns hidden in complex electroencephalographic (EEG) data. Conditions such as epilepsy, Alzheimer’s disease, dementia, and autism exhibit distinct alterations in neural oscillations and connectivity, which remain difficult to interpret in real time; therefore, this study proposes an interactive interface for intuitive exploration and analysis of disease-specific EEG dynamics. The system integrates classical signal processing techniques and computational modeling to extract spectral features, inter-electrode coherence, and spatial activation patterns, which are visualized through spectrograms, topographic maps, and connectivity graphs that update continuously. In addition, a web-based platform is incorporated to enable clinicians and technicians to store and manage patient information, including diagnosis, severity level, number of recordings, sampling frequency, recording duration, and acquisition dates, supporting structured data organization and longitudinal monitoring. The results demonstrate that the interface captures meaningful differences between disorders, with epileptic patterns showing strong synchronization and burst activity, while neurodegenerative conditions exhibit spectral slowing and reduced connectivity. Overall, the proposed framework provides an effective and accessible tool for EEG visualization, combining interactive analysis with clinical data management to support research, education, and potential clinical applications. Full article

Suspended sediment concentration (SSC) is a key indicator of river sediment transport processes and water environmental change. For medium-width rivers, continuous-reach SSC monitoring remains constrained by the spatial discontinuity of station observations and the temporal or consistency limitations of single-source satellite imagery. To improve multi-year SSC characterization in the middle and lower reaches of the Liaohe River, this study integrated Harmonized Landsat and Sentinel-2 (HLS) surface reflectance imagery from 2016 to 2022 with SSC observations from five hydrological stations and developed a random forest retrieval model using multi-band reflectance and sediment-related spectral features. The trained model was applied to valid HLS images to examine SSC spatial distribution, interannual variation, and inter-station reach differences. The model achieved a test-set R² of 0.641, an RMSE of 0.083 kg·m⁻³, and an MAE of 0.067 kg·m⁻³. The median composite of 52 retrieval images showed a lower SSC in the Tieling–Mahushan and Mahushan–Pinganbao reaches and a higher SSC in the Pinganbao–Liaozhong and Liaozhong–Liujianfang reaches. SSC was generally higher in 2016 and 2022 and lower in 2018. These findings indicate that HLS-based retrieval can support continuous-reach SSC monitoring and regional water–sediment dynamic assessment in medium-width rivers, although the accurate quantification of extreme high-SSC events still requires additional in situ samples and higher-frequency observations. Full article

Long and narrow shielded confined spaces, represented by traffic tunnels and underground utility tunnels, constitute critical application scenarios for indoor and underground positioning services. Despite their relatively simple geometric configurations, such environments suffer from severe spatial distortion of geometric dilution of precision (GDOP). Coupled with pervasive low-elevation signal propagation and intensive multipath reflection effects, conventional BeiDou Navigation Satellite System (BDS) positioning services are unable to provide continuous and reliable coverage in these scenarios. To date, existing research on high-precision pseudolite positioning for narrow confined spaces remains largely confined to theoretical analysis and laboratory experimental verification, while systematic studies on application-oriented signal atlas feature network design are significantly insufficient, forming a prominent gap that restricts the practical engineering deployment of relevant technologies. To address the aforementioned technical bottlenecks, this paper proposes a novel BDS pseudolite signal atlas network design method to improve the continuity, stability and comprehensive positioning performance in spatially distorted narrow shielded environments. Field vehicular tests were carried out in actual engineering tunnels and underground utility tunnels to systematically analyze the variation characteristics of raw BDS pseudolite observation data, including pseudorange, carrier phase, carrier-to-noise ratio (C/N₀) and Doppler shift. The test results verified that kinematic Doppler parameters exhibited outstanding stability in complex shielded environments with strong multipath interference. On this basis, a spatial feature model based on kinematic Doppler measurements was constructed, and wavelet denoising technology was adopted to extract effective typical spatial feature parameters. Combined with the deterministic one-to-one mapping relationship between Doppler peak characteristics and spatial positions, a multi-peak kinematic Doppler atlas was established, which eliminates the dependence on pre-deployment data collection, dedicated database construction and offline model training. Furthermore, comprehensively considering multi-dimensional constraints such as spatial environment scale, carrier dynamic characteristics and terminal output rate, the atlas network scheme was optimized to achieve a balanced trade-off among positioning detection accuracy, absolute positioning precision and suppression of the pseudolite near-far effect. Comparative experimental results demonstrate that the proposed BDS pseudolite atlas network effectively resolves the inherent GNSS positioning difficulty in long and narrow shielded spaces. Benefiting from the rational spectral peak configuration strategy, the system can satisfy the continuous and stable positioning requirements of multiple carrier types including motor vehicles and railway locomotives under variable motion speeds and terminal output rates. This study provides a robust and feasible technical solution for high-precision BDS positioning services in long and narrow shielded confined spaces, and holds favorable engineering application prospects for underground navigation scenarios. Full article

Hyperspectral remote sensing enables precise identification of alteration mineral through spectral–image integration and high-resolution capabilities. However, vegetation interference significantly hinders the extraction of alteration information in vegetated areas, thereby posing challenges to the reliable identification of alteration minerals. This study employs GF-5B satellite AHSI imagery acquired in the Huzhou region of Zhejiang Province, China, to address this challenge via a novel Zonal Adaptive Vegetation Suppression Technique (ZAVST). By constructing segmented statistical models that links reflectance characteristics across multiple spectral bands to NDVI values, ZAVST demonstrates an enhanced capability to mitigate vegetation obscuration effects on subsurface lithological features while substantially improving the identification of subtle spectral signatures characteristic of mineralization. Results reveal distinct spatial patterns: Fe-bearing alteration minerals (hematite, pyrite) align along NE-trending faults and volcanic basin margins; Al-OH alterations (montmorillonite, kaolinite) cluster near intrusive contacts; Mg-OH alterations (chlorite, epidote) occur at interfaces between carbonate sequences and concealed intrusions. Composite alteration anomalies exhibiting stacked mineral signatures (up to four distinct types) were identified across the region, demonstrating a strong spatial correlation with known mineralization centers. By integrating alteration zonation, structural lineaments, stratigraphy, geochemical anomalies, and orebody records, this study delineated four priority targets: Lijiaxiang Town, eastern Meixi Town, Miaoxi Town, and the central Moganshan Volcanic Basin. Full article

Synthetic aperture radar (SAR) object detection has long been affected by spatial speckle interference, spectral energy imbalance, and structural bias in cross-scale feature fusion. In this article, we propose the Multi-Domain Interference-Suppressed Detection Transformer (MDIS-DETR), a unified multi-domain interference-suppressed detection framework built on the Real-Time Detection Transformer (RT-DETR) architecture. Specifically, spatial-domain interference is suppressed by learnable fusion of complementary denoising responses at the input stage. Furthermore, frequency-domain interference is suppressed by polarization-guided attention together with adaptive frequency refinement within the encoder. In addition, structural-domain interference is suppressed by non-sequential cross-scale interaction to enhance multi-scale consistency. Extensive experiments on multiple SAR benchmarks demonstrate that MDIS-DETR establishes state-of-the-art (SOTA) performance across datasets. Notably, on SARDet-100K, currently the largest SAR detection dataset with a scale comparable to the Common Objects in Context (COCO) dataset, it achieves 58.82% mAP, surpassing the RT-DETR baseline by 4.58%. Full article

Rapid and reliable flood segmentation utilizing optical remote-sensing imagery is critical for effective flood disaster response and risk assessment. Nevertheless, current models frequently struggle with imprecise boundary delineation and fragmented predictions in complex environments, especially where floodwater displays high spectral variability and closely resembles shadows, dark pavements, or wet soil. To overcome these challenges, we introduce FloodSeg, an innovative Mamba-CNN encoder–decoder network incorporating two lightweight yet highly effective components: a Shift module and a sequence-shuffle module. The spatial Shift module leverages spatially shifted feature aggregation to fortify boundary-aware representations, thereby ensuring the continuity of inundation contours even under varying illumination and cluttered backgrounds. Meanwhile, the sequence-shuffle module reorganizes multi-scale features via sequence-wise mixing and cross-regional interaction, significantly enhancing long-range dependency modeling. This facilitates the generation of globally consistent flood masks while mitigating local overfitting to dataset-specific textures. Evaluated on the Kaggle and FloodNet benchmark datasets, FloodSeg achieves outstanding mIoU scores of 81.85% and 91.21%, respectively. By outperforming various state-of-the-art CNN-, Transformer-, and Mamba-based baselines, our model demonstrates a superior accuracy-efficiency trade-off. These results substantiate that FloodSeg significantly advances boundary recognition and overall segmentation completeness, establishing it as a robust and practical solution for real-world remote-sensing flood mapping applications. Full article

Hyperspectral image (HSI) stripe noise removal is essential for downstream interpretation tasks. However, most existing methods exhibit incomplete joint modeling of spatial structures and inter-band spectral correlations, lack direction-aware modeling for stripe noise, and lack differentiated processing of high- and low-frequency components. To tackle these limitations, we propose a spatial-spectral attention-enhanced multi-level wavelet-informed network (SAMWNet). Its dual-branch module extracts spatial and spatial-spectral features from each band and its adjacent bands. Afterward, a discrete wavelet-informed progressive denoising (MDWPD) module conducts multi-level Haar wavelet decomposition and progressive reconstruction. Within this module, the low-frequency hybrid enhancement (LFHE) module preserves low-frequency spectral structures, while the high-frequency enhancement (HFME) module suppresses directional stripe artifacts in high-frequency subbands. We further adopt a composite loss function to balance pixel fidelity, noise estimation, structural similarity, and spectral consistency. Experimental results on simulated and real-world HSIs demonstrate that SAMWNet achieves competitive or superior performance compared with several representative HSI denoising methods. Full article

Wheat is a major staple crop, and storage mold growth poses a severe threat to grain safety and quality stability. Natural mold development in stored wheat exhibits subtle, localized, and highly heterogeneous characteristics. Existing unimodal methods and global fusion approaches generally suffer from insufficient local feature sensitivity, hindering fine-grained mold severity grading. To address this limitation, we propose a Mask-Guided Fine-Grained Fusion Network, a weakly supervised framework based on local RGB–HSI fusion. This framework employs a dynamic parallel A/B experimental design to construct time-matched proxy labels via weakly supervised learning. A standardized preprocessing pipeline including single-kernel extraction, foreground segmentation, and cross-modal registration is established to resolve RGB–HSI spatial misalignment, ensuring physical-level spatial consistency of multimodal features. The model incorporates a Foreground-Aware Spectral Recalibration (FASR) module to suppress background noise, a Mask-Guided Dilated Cross-modal Local Attention (MDCLA) mechanism to establish fine-grained local mappings between RGB visual phenotypes and hyperspectral responses, and a sample-level adaptive fusion strategy to dynamically weight features by modal reliability, enhancing representation of complex samples across all mold stages. Experiments show that the Mask-Guided Fine-Grained Fusion Network achieves 0.9689 classification accuracy, 0.9698 Macro-F1 score, and 0.0593 Mean Absolute Error (MAE), significantly outperforming state-of-the-art unimodal deep models and global attention fusion baselines. This work provides a proof-of-principle framework for fine-grained non-destructive mold risk assessment in stored wheat. Full article

Accurate land use/land cover (LULC) mapping from remotely sensed data remains challenging in arid regions, particularly for spatial and seasonal generalization. This work proposes a novel exclude-one-city-out (EOCO) framework based on machine learning (ML) to achieve LULC generalization across summer and winter in arid environments. Four cities in Saudi Arabia witnessing rapid urban growth were selected: Riyadh, Madinah, Jeddah, and Dammam. The ML models were trained on three cities and tested on the unseen city. Sentinel-2 surface reflectance data for the visible (Blue, Green, and Red) and near-infrared bands (NIR, SWIR1, and SWIR2) were used. Spectral indices, texture features, and topographical data were used to form five feature sets, which were utilized as inputs for four ML algorithms: random forest, support vector machine, classification and regression trees, and K-nearest neighbors. Statistical tests (Friedman, Kendall’s W, and Wilcoxon signed rank) were conducted to assess differences across ML models, feature sets, and seasons. The random forest model consistently outperformed other models across the five feature sets, while the spectral texture and combined feature sets outperformed other feature combinations. Significant differences in feature importance were observed across cities and seasons for spectral texture during summer and winter (p-values: 1.25 × 10⁻⁴ and 9.2 × 10⁻⁵, respectively), with strong agreement (Kendall’s W = 0.9212 and 0.9424). The findings can support urban environmental monitoring in arid regions, contributing to sustainable urban development. Full article

Frequent seasonal phase transitions in cold and arid lakes require different remote sensing indices for frozen and open-water periods, complicating the use of traditional empirical indices for automated monitoring. To address this challenge, this study proposes an intelligent indexing framework integrating the heuristic reasoning of Large Language Models (LLMs) with Random Forest (RF) feature selection. Leveraging the Google Earth Engine (GEE) and Landsat 8 data from Ulansuhai Lake, five LLMs such as Gemini and ERNIE were employed to generate candidate spectral indices based on typical sample spectra. Optimal band combinations were identified via RF importance, and Land Surface Temperature (LST) was incorporated as a physical constraint for unified cross-seasonal classification and determine the optimal threshold. Results show that the LLM-derived ERNIE-WISI and Gemini-WISI exhibit high robustness. During the freezing period, ERNIE-WISI significantly outperformed other indices, achieving an Overall Accuracy (OA) of 89% and a Kappa of 0.86. Spatially, it yielded snow and ice mapping with clear textures and low commission errors. During the non-freezing period, ERNIE-WISI achieved an OA of 95% with a Kappa of 0.84. While Gemini-WISI achieved an OA of 94% with a Kappa of 0.80, performing comparably to MNDWI. Notably, ERNIE-WISI effectively suppressed background interference in complex landscapes like narrow channels and aquaculture areas, maintaining high geometric fidelity and spatial continuity. A key advantage of ERNIE-WISI is its consistent performance without seasonal threshold adjustments. Aligned with the AI for Science paradigm, this methodology bridges AI-driven heuristic discovery and physical remote sensing, offering a robust, transferable solution for long-term dynamic lake monitoring in extreme environments, thereby facilitating sustainable water management. Full article

Tropical rainforests are characterized by high species diversity and complex canopy structure, making accurate forest stand classification important for ecosystem assessment, biodiversity monitoring, and forest carbon estimation. However, single-source remote sensing data lacks sufficient discrimination ability to address the issue of spectral similarity among classes, and conventional convolutional neural networks often struggle to extract discriminative features and integrate heterogeneous data in highly complex forests. To address these challenges, this study developed a dual-branch deep learning framework that integrates DenseNet and ConvNeXt for classification in Hainan Tropical Rainforest National Park. The framework combines sub-meter Google Earth imagery to capture spatial–textural detail with multi-temporal Sentinel-2 imagery to represent phenological variation. The results showed that multi-temporal Sentinel-2 data outperformed single-date imagery by capturing phenological patterns, and that the fusion of high-resolution spatial information and multi-temporal spectral information yielded higher accuracy than either data source alone. The dual-branch model achieved an overall accuracy of 94.47% and a Kappa coefficient of 0.94, outperforming all benchmark models. These findings indicate that branch-specific feature extraction and adaptive fusion can improve fine-scale classification in complex tropical rainforest environments. The proposed framework provides a practical approach for fine-scale forest stand mapping and may support biodiversity monitoring, ecological assessment, and sustainable forest management. Full article

Rapid identification of maize waterlogging is essential for post-disaster agricultural assessment, but most existing methods rely on multi-temporal imagery that is often unavailable immediately after extreme rainfall events. This study proposes SAB-DeepLabV3+, a semantic segmentation model for mapping waterlogging-affected maize from single-date multispectral imagery within pre-extracted maize planting areas. Built on DeepLabV3+, the model integrates three task-specific modules: a Spectral-Spatial Information Enhancement Module to improve feature discrimination under spectral mixing, an Adaptive Multi-Scale Pooling Module to capture heterogeneous patch sizes, and a Boundary Enhancement Module to refine transition zones. A pixel-level dataset containing 12,198 image patches was constructed from 62 multispectral scenes collected across five major maize-producing cities in Heilongjiang Province, China, during 2022–2024. On the test set, SAB-DeepLabV3+ achieved a waterlogged-class IoU of 68.30%, mIoU of 80.37%, mF1 of 88.62%, and OA of 93.49%, outperforming DeepLabV3+. Leave-one-city-out evaluation further produced an average mIoU of 76.56% and a waterlogged-class IoU of 63.45%. These results indicate that single-date high-resolution multispectral imagery can support rapid and reliable maize waterlogging mapping. Full article

Pegmatite dikes are important prospecting indicators for rare-metal deposits, whereas traditional methods for pegmatite dike identification are constrained by the limited capability of human visual interpretation to capture information from remote sensing imagery, resulting in low identification accuracy and efficiency. In recent years, global research on semantic segmentation of different surface features and remote sensing-based mineral exploration using deep learning methods and high-resolution remote sensing imagery has made significant progress; however, studies on surface-exposed geological bodies such as pegmatite dikes remain highly insufficient. To address the key problem of efficiently identifying pegmatite dikes in remote sensing imagery, this study proposes an improved model based on UNet++, termed GAD-UNet++. In the field of remote sensing geology, this study constructed a pegmatite dike semantic segmentation dataset based on high-resolution RGB imagery by using 0.66 m RGB imagery for visual delineation and ZY1F hyperspectral data for spectral constraint and label refinement; on this basis, semantic segmentation of surface pegmatite dikes in the Yilanlike area of the South Tianshan Mountains, Xinjiang, was conducted using RGB remote sensing image patches as model input. Specifically, because pegmatite dikes are small targets characterized by slender structures, indistinct boundaries, and sparse regional distribution, this study introduced a lightweight feature extraction structure (GhostNetV2) and a long-range dependency attention module (DFC) at the encoder stage, and further incorporated the Coordinate Attention module (CA) to enhance spatial localization and boundary representation of the targets. Finally, focal cross-entropy loss and a deep supervision strategy were adopted to improve the accuracy of semantic information extraction for pegmatite dikes, as well as the training stability and segmentation accuracy under class-imbalance conditions. The results show that the proposed model achieved an mIoU of 93.11% and an F1-score of 94.95% on the test set. Compared with existing semantic segmentation models, the proposed model achieved superior performance in both identification accuracy and computational efficiency for pegmatite dikes. In addition, this study delineated 18 potential pegmatite dike enrichment zones in the Yilanlike area, providing technical support for remote sensing-based rare-metal prospecting and geological interpretation in the study area. Full article

Cross-scene Hyperspectral Image (HSI) classification faces substantial domain shifts caused by sensor heterogeneity, acquisition variation, and scene diversity. While benchmark annotations are assigned to individual center pixels, local patches often contain implicit multi-label semantics due to spectral mixing and spatial overlap. This mismatch distorts prediction structure, exacerbates generalization errors, and limits the effectiveness of standard domain generalization (DG) techniques focused solely on feature or prediction invariance. We propose Class-Aware Semantic Calibration (CASC), a systematic semantic structure calibration framework that addresses three complementary distortions induced by mismatched patch supervision: (i) Balance corrects class frequency bias via reweighted supervision; (ii) Separability enhances boundary decision stability through margin-based logit calibration; and (iii) Independence reduces domain-specific spurious co-occurrence via prediction covariance decorrelation. To preserve calibrated semantics under pseudo-source shift, we further introduce a complementary DualAlign (DA) module, which jointly aligns feature statistics and prediction distributions, enforcing consistency at both representation and semantic levels. Extensive experiments on three cross-scene benchmarks (Houston, Pavia, and WHU-Hi) demonstrate that CASC-DA consistently improves performance over strong baselines, achieving an average gain of 3.0% in overall accuracy and 4.9% in Kappa coefficient compared with the best-performing baseline on each dataset. These results underscore the importance of semantic structure calibration for domain-generalized HSI classification. Full article