IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society最新文献

{"title":"Adaptive Sample Allocation for SAR Ship Detection Based on Scale-Sensitive Wasserstein Distance","authors":"Shibo Chang;Xiongjun Fu;Jian Dong;Weidong Hu;Weihua Yu","doi":"10.1109/LGRS.2025.3597146","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3597146","url":null,"abstract":"Deep learning (DL) based synthetic aperture radar (SAR) imagery ship detection is challenged by multiscale ships on the identical SAR image, which inevitably leads to insufficient and low-quality positive samples during training and ultimately degrades detection performance. To address this issue, we propose a Scale-Sensitive Adaptive Sample Allocation Strategy (SSA-SAS) for SAR ship detection. SSA-SAS ranks candidate boxes using a unified score that integrates a scale-sensitive Wasserstein distance (SSWD), a shape cost, and classification confidence. SSWD serves as the core regression metric, enabling adaptive tolerance to positional offsets based on object scale. Meanwhile, the shape cost introduces morphological priors to guide early-stage optimization. These components jointly enhance the quantity and quality of selected positive samples throughout training. Experimental results show that SSA-SAS improves average precision (AP) by up to 2.6% on the high-resolution SAR images dataset for ship detection and instance segmentation (HRSID) dataset and 1.4% on the SAR ship detection dataset (SSDD), while accelerating network convergence by approximately 5.0%.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":4.4,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144868193","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient Water Body Detection Based on Knowledge Distillation for SAR Imagery","authors":"Jinze Zhu;Shibao Li;Yunwu Zhang;Menglong Liu;Jiaxin Chen","doi":"10.1109/LGRS.2025.3597141","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3597141","url":null,"abstract":"Synthetic aperture radar (SAR) is widely used for water body detection due to its efficiency and ability to operate in all weather conditions. However, its scattering properties and single-polarization limitations pose challenges for data extraction and reduce the accuracy of water body detection algorithms. To mitigate this limitation, recent studies have focused on transforming SAR datasets into electro-optical (EO) image modalities through cross-modal translation models, aiming to enhance multispectral feature interpretability. However, such transformation frameworks require substantial computational power, which compromises the real-time processing capabilities critical for rapid disaster response, such as a flood. In this letter, we propose a lightweight SAR water body detection framework that integrates knowledge distillation and channel attention. A teacher network trained on rich EO data guides an SAR-specific student model, with both employing attention branches. The student’s attention is supervised by the teacher to enhance SAR feature extraction via attention-aligned distillation. Evaluated on the Sen1Floods11 benchmark dataset, our experimental results outperform the baseline model by 3.5% in intersection over union (IoU).","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":4.4,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144914172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Operational Assessment of Side-Scan Sonar Data Applied to Naval Mine Detection Using an Automatic Target Recognition Algorithm","authors":"Camilla Caricchio;Luis Felipe Mendonça;André T. C. Lima;Carlos A. D. Lentini","doi":"10.1109/LGRS.2025.3596852","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3596852","url":null,"abstract":"Mine warfare (MW) and mine countermeasures (MCMs) have become strategic options to ensure national sovereignty and the safety of maritime commercial routes, which is the primary logistics system for international trade. As an asymmetric weapon, locating and neutralizing a naval mine poses a significant challenge for the world’s navies. In this context, this work proposes an object detection model based on you only look once, version 11 (YOLOv11) for automatic and real-time detection of naval mines in harbor areas using side-scan sonar (SSS) data. The main objective of this tool is to apply it to unmanned maritime vehicles (UMVs) to enhance the mine detection efficiency during minehunting operations. Second, this study aims to evaluate the effects of operational parameters, oceanographic and meteorological conditions on the SSS data quality for naval mine detection. All the data used to train the neural network were real and obtained in a test area, mimicking a port area, a strategic environment in the context of MW. The model performed with satisfactory statistical results (mAP@0.5: 0.84, P: 0.93, R: 0.83, and F1 score: 0.88). Based on the results provided in this study, the 0.70 confidence level can be safely used in future operational inferences using this customized model. From the operational evaluation of SSS data quality, the ideal condition for data acquisition is using an intermediary range and high-frequency sonars with calm seas and low speeds. Despite the recent advancements in the field of machine learning, it is unlikely that neural networks will fully replace human operators in MCM missions in the short to medium term. However, they serve as a valuable tool for decision support, enabling rapid analysis of large datasets and filtering information to present only the most relevant data to human analysts, such as potential sea mines. When embedded in UMV, this technology mitigates risks to human life and enables operators to focus on verifying real targets, thereby enhancing the effectiveness of MCM operations.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":4.4,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144853439","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sea Clutter Influencing Factors Analysis and Parameter Estimation Based on Oceanographic Observations","authors":"Xian Yu;Yubing Han;Binyun Yan;Weixing Sheng","doi":"10.1109/LGRS.2025.3596590","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3596590","url":null,"abstract":"Accurate and robust sea clutter modeling and parameter estimation are foundational for target detection. Traditional modeling methods rely on measured data, while clutter modeling based on radar settings and oceanographic observations is an alternative. This letter leverages high-resolution sea clutter data from the Sea-Detecting Radar Data-Sharing Program (SDRDSP) to address this challenge. Three distribution types, which are generalized Pareto distribution (GPD), K distribution, and compound-Gaussian model with inverse Gaussian (CGIG), are considered. Using random forest (RF), we identify the most discriminative factors for distribution type classification: range and azimuth resolution cell (RARC), grazing angle, wave speed, wind speed, and significant wave height (SWH). Building on this, a stacking ensemble learning framework is proposed to effectively regress the shape and scale parameters from these optimized input features. Experiments validate the effectiveness of the proposed approach in distribution type classification and parameter estimation.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":4.4,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144914171","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"EnergyFormer: Energy Attention With Fourier Embedding for Hyperspectral Image Classification","authors":"Saad Sohail;Muhammad Usama;Usman Ghous;Manuel Mazzara;Salvatore Distefano;Muhammad Ahmad","doi":"10.1109/LGRS.2025.3596629","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3596629","url":null,"abstract":"Hyperspectral images (HSIs) capture detailed spectral–spatial information across hundreds of contiguous bands, enabling precise material identification in domains such as environmental monitoring, agriculture, and urban analysis. However, the high dimensionality and spectral variability inherent to HSIs present significant challenges for effective feature extraction and classification. This letter introduces EnergyFormer (EF), a transformer-based framework designed to overcome these limitations through three key innovations: 1) multihead energy attention (MHEA), which formulates an energy optimization mechanism to selectively enhance discriminative spectral–spatial features; 2) Fourier positional embedding (FoPE), which adaptively models long-range spectral and spatial dependencies; and 3) enhanced convolutional block attention module (ECBAM), which emphasizes informative wavelength bands and spatial structures for robust representation learning. Extensive experiments on the WHU-Hi-HanChuan, Salinas, and Pavia University datasets demonstrate that EF achieves superior classification performance with overall accuracies of 99.28%, 98.63%, and 98.72%, respectively, outperforming leading CNN-, transformer-, and Mamba-based models.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":4.4,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144853484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Geometric Correction of Bistatic SAR Moon Mapping Results Based on the RPC Model","authors":"Yan Yin;Jinghai Sun;Lijia Huang;Jingxing Zhu;Peng Jiang;Chibiao Ding","doi":"10.1109/LGRS.2025.3596794","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3596794","url":null,"abstract":"The bistatic synthetic aperture radar (SAR) system based on the Five-Hundred-Meter Aperture Spherical Radio Telescope (FAST) has successfully acquired multiple high-quality Moon surface images. However, the system errors introduced by noncooperative transmitting and receiving radars prevent traditional geometric correction methods based on geographic data and radar parameters from accurately transforming delay-Doppler format images into geographic coordinates, making it impossible to further utilize these images for scientific applications. In this letter, a novel geometric correction method is proposed for transforming delay-Doppler images to the Moon’s geographic coordinate images. This approach utilizes a geometric positioning model to generate ideal delay-Doppler coordinates corresponding to the Moon’s geographic coordinates. These coordinate pairs are used to fit the rational polynomial coefficient (RPC). Subsequently, based on the RPC model, localization offsets are corrected through an affine transformation and selective coefficient optimization. An optical-SAR image registration method is used to determine the localization offsets and evaluate the reliability of the geometric correction method. We demonstrate this approach using Moon SAR images obtained by the bistatic SAR system based on FAST and other transmitting radars. This method can effectively integrate multisource Moon SAR data to address specific scientific challenges.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":4.4,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144868194","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DMSDA-YOLO: Dynamic Multiscale Dilated Attention for Remote Sensing Object Detection","authors":"Zhenghua Huang;Zijian Xu;Xi Li;Yaozong Zhang;Yu Shi;Qian Li;Hao Fang","doi":"10.1109/LGRS.2025.3596809","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3596809","url":null,"abstract":"It is an extremely challenging task to detect multiscale targets (especially small objects) in remote sensing (RS) images with complex backgrounds. This letter develops a novel RS object detection model, namely dynamic multiscale dilated attention based on YOLOv5 (DMSDA-YOLO), of which the key improvements include: one is that, in the backbone, a multiscale dilated attention fusion module (MDAFM) is proposed to capture multiscale feature information and a coordinate anchor attention (CAA) mechanism is incorporated to increase the focus on target regions while suppressing background interference. The other is that a spatial attention pyramid neck network is proposed to improve its feature fusion capability while a dynamic attention-aware feature extraction module (DAFEM) is introduced to enhance the network’s adaptability to multiscale targets in the neck. Objective and subjective results of experiments on the DIOR, HRRSD, and NWPU VHR-10 datasets demonstrate that our DMSDA-YOLO outperforms existing state-of-the-art object detection approaches in detecting multiscale targets under complex backgrounds, and its competitive computational complexity is beneficial for its extensive application.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":4.4,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144880425","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Examining the Derived Sea Wave Heights From ICESat-2 Weak Beams: A Case Study in Marginal Seas","authors":"Zhibiao Zhou;Jian Yang;Yue Ma;Qi Liu;Yue Song;Song Li","doi":"10.1109/LGRS.2025.3596699","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3596699","url":null,"abstract":"The Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2) carries the new generation spaceborne photon-counting lidar, Advanced Topographic Laser Altimeter System (ATLAS). ICESat-2/ATLAS has an excellent performance for obtaining precise geometric surface profiles of land and oceans, by which the surface parameters such as the significant wave height (SWH) over oceans can be further obtained. As the strong beams have better data quality, they are currently used to obtain the sea surface parameters. The weak beams could double the spatial coverage area if they can also be successfully used. However, this potential is constrained by the lower signal-to-noise ratio (SNR) of weak beams. To exploit the performance of weak beams, this study proposes a method to extract sea surface signal photons, which are further accumulated to calculate the SWHs. This study explores the effect of the data processing window length on the result of the denoising algorithm and how many sea surface signal photons should be accumulated to estimate the reliable SWHs with ICESat-2 weak beams. The calculated SWH shows good agreement with ECMWF reanalysis 5 (ERA5) data, with the root mean square error (RMSE) under 0.3 m. The method proposed in this study enables the acquisition of SWH values in certain regions where no ATL12 official data are available.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":4.4,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144880630","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient JPEG-AI Image Coding for Remote Sensing Semantic Segmentation","authors":"Junxi Zhang;Xiang Pan;Zhenzhong Chen;Shan Liu","doi":"10.1109/LGRS.2025.3596235","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3596235","url":null,"abstract":"Efficient image compression is crucial for remote sensing (RS) satellite systems, as it determines the performance of machine vision applications analyzing the downlinked image data at ground stations. However, existing conventional or learning-based image compression approaches exhibit limitations in either high complexity or undesirable vision task performance. This letter investigates an efficient neural image compression standard, JPEG-AI-based self-supervised RS image compression approach, and SS-JPEG-AI, for semantic segmentation tasks. Our approach maintains the low-complexity advantages of JPEG-AI while incorporating: 1) a computationally efficient transformer-based attention mechanism that enhances reconstruction quality without increasing encoder complexity for onboard systems and 2) a contrastive learning strategy that improves feature discriminability and sharpens intercategory decision boundaries for segmentation tasks. Compared to the state-of-the-art image compression methods, SS-JPEG-AI achieves better Bjøntegaard delta-rate (BD-rate) performance across the mean intersection over union (mIoU) and mean F-score (mFscore) while maintaining up to <inline-formula> <tex-math>$30times $ </tex-math></inline-formula> smaller computational complexity.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":4.4,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144853485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Adaptive Electromagnetic Analysis via Non-Euclidean Manifold Learning for Atmospheric Precipitation Understanding","authors":"Tian Fu;Tianliang Yao;Haoyu Wang;Bin Chen","doi":"10.1109/LGRS.2025.3596318","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3596318","url":null,"abstract":"The advent of dual-polarization meteorological sensing systems has revolutionized our capacity to comprehend atmospheric precipitation dynamics through electromagnetic signal analysis. However, the intricate nonlinear relationships within high-dimensional polarimetric signatures present formidable challenges in extracting actionable intelligence for meteorological multimedia applications. This paper presents HyperSpectral-M, a computational framework that enhances polarimetric signal interpretation through systematic manifold learning approaches in non-Euclidean spaces, enabling more precise analysis of complex atmospheric phenomena. The proposed HyperSpectral-M framework addresses the limitations of the existing methods by incorporating two key innovations: a signal disentanglement mechanism (SDM) and a physics-constrained reconstruction paradigm (PCRP). The disentanglement mechanism employs quaternion-based geodesic flow mapping coupled with adaptive spectral decomposition (SD) to project polarimetric signatures onto lower dimensional manifolds while preserving critical microphysical properties. This is augmented by a multiscale differential geometry analyzer that captures intricate spatiotemporal correlations across varying atmospheric conditions. The reconstruction paradigm leverages adversarial manifold alignment with structured probabilistic inference to synthesize high-fidelity radar representations while maintaining electromagnetic consistency constraints. HyperSpectral-M demonstrates significant real-world impact on meteorological applications by improving precipitation nowcasting accuracy by 15–20% compared to operational methods, enabling more timely and accurate flood warnings. Field validation with emergency management agencies shows that reduces false alarm rates by 30–40% while increasing lead time for severe weather warnings by 15–30 min.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"22 ","pages":"1-5"},"PeriodicalIF":4.4,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144843028","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}