IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing最新文献

{"title":"Enhancing, Refining, and Fusing: Towards Robust Multiscale and Dense Ship Detection","authors":"Congxia Zhao;Xiongjun Fu;Jian Dong;Shen Cao;Chunyan Zhang","doi":"10.1109/JSTARS.2025.3556893","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3556893","url":null,"abstract":"Synthetic aperture radar (SAR) imaging, celebrated for its high resolution, all-weather capability, and day-night operability, is indispensable for maritime applications. However, ship detection in SAR imagery faces significant challenges, including complex backgrounds, densely arranged targets, and large scale variations. To address these issues, we propose a novel framework, Center-Aware SAR Ship Detector (CASS-Det), designed for robust multiscale and densely packed ship detection. CASS-Det integrates three key innovations: 1) a center enhancement module (CEM) that employs rotational convolution to emphasize ship centers, improving localization while suppressing background interference; 2) a neighbor attention module that leverages cross-layer dependencies to refine ship boundaries in densely populated scenes; and 3) a cross-connected feature pyramid network (CC-FPN) that enhances multiscale feature fusion by integrating shallow and deep features. The proposed model achieves mean Average Precision of 99.2%, 93.1%, and 82.1% on the SSDD, HRSID, and LS-SSDD datasets, surpassing the second-best methods by 1.2%, 1.8%, and 1.8%, respectively, which demonstrates its effectiveness.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"9919-9933"},"PeriodicalIF":4.7,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10946877","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143860873","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Designing a Classifier for Active Fire Detection From Multispectral Satellite Imagery Using Neural Architecture Search","authors":"Amber Cassimon;Phil Reiter;Siegfried Mercelis;Kevin Mets","doi":"10.1109/JSTARS.2025.3556550","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3556550","url":null,"abstract":"Wildfires are becoming increasingly devastating, and detecting them early is essential to containing them. Deep learning-based wildfire detection systems have increased in complexity dramatically in recent years, and in order to manage this added complexity, techniques have been proposed to automate the design of neural network architectures. Such techniques are usually referred to as neural architecture search (NAS). This article showcases the use of a reinforcement learning-based neural architecture search (NAS) agent to design a small neural network to perform active fire detection on multispectral satellite imagery. Specifically, we aim to automatically design a neural network that can determine if a single multispectral pixel is a part of a fire, and do so within the constraints of a low earth orbit nanosatellite with a limited power budget, to facilitate on-board processing of sensor data. A regression model that predicts the F1 score obtained by a particular architecture following quantization is used as a reward function. This model is trained on the classification performance statistics of a sample of neural network architectures. Besides the F1 score, we also include the total number of parameters in our reward function to limit the size of the designed model. Finally, we deployed the best neural network to the Google Coral Micro Dev Board and evaluated its inference latency and power consumption. This neural network consists of 1716 parameters, takes on average 984 <inline-formula><tex-math>$mu$</tex-math></inline-formula>s to inference, and consumes around 800 mW to perform inference. These results show that our approach can be applied to new problems.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"10204-10224"},"PeriodicalIF":4.7,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10946665","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143860841","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Trackwise Prediction of GNSS-R Delay–Doppler Maps With DDM-PredRNN Network","authors":"Weichen Sun;Xiaochen Wang;Bing Han;Dongkai Yang","doi":"10.1109/JSTARS.2025.3555623","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3555623","url":null,"abstract":"Global navigation satellite system reflectometry (GNSS-R) is an emerging Earth observation method that utilizes reflection signals from navigation satellites for remote sensing of physical parameters, particularly in detecting ocean wind speed. However, challenges, such as low signal-to-noise ratio (SNR) and insufficient gain of the receiving antenna have hindered the acquisition of reliable observation data in many areas. With the high temporal resolution offered by GNSS-R, the delayed-Doppler map (DDM) of radar echo images reveals inherent spatial continuity and temporal correlation across sequential intervals. This article introduces an advanced deep learning framework DDM-PredRNN for predicting DDM in previously unobserved regions. Employing a spatiotemporal long short-term memory network with a PredRNN architecture, this model integrates axial attention to capture and leverage the delay/Doppler features embedded within the DDM, thereby enhancing the extraction of complex spatiotemporal characteristics. Experiments utilizing cyclone GNSS data illustrate that the proposed method can effectively predict unknown DDMs along the GPS motion track within a 20-step timeframe. The results indicate that the root mean square error of the DDM prediction is 1.32 dB, the mean absolute error is 0.82 dB, and the structural similarity is 0.71. This approach not only effectively addresses the gaps in ocean wind observation data due to poor GNSS-R data quality but also fills the trackwise blind zones of GNSS-R. Furthermore, the predicted DDM accurately reflects trends in the NBRCS near specular point, providing a new perspective for GNSS-R retrieval forecasting of ocean wind.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"9837-9849"},"PeriodicalIF":4.7,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10945374","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143850891","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing the SMAP Level-4 Carbon Product Over the Arctic and Subarctic Zones","authors":"Rémi Madelon;John S. Kimball;K. Arthur Endsley;Gabriëlle J. M. De Lannoy;Oliver Sonnentag;Haley Alcock;Matteo Detto;Anna M. Virkkala;Brendan M. Rogers;Jennifer D. Watts;Alex Mavrovic;Scott N. Williamson;Elyn Humphreys;Andreas Colliander;Arnaud Mialon;Alexandre Roy","doi":"10.1109/JSTARS.2025.3555850","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3555850","url":null,"abstract":"The soil moisture active passive (SMAP) satellite mission distributes a product of CO<inline-formula><tex-math>$_{text{2}}$</tex-math></inline-formula> flux estimates (SPL4CMDL) derived from a terrestrial carbon flux model, in which SMAP brightness temperatures are assimilated to update soil moisture (SM) and constrain the carbon cyclemodeling. While the SPL4CMDL product has demonstrated promising performance across the continental USA and Australia, a detailed assessment over the arctic and subarctic zones (ASZ) is still missing. In this study, SPL4CMDL net ecosystem exchange (NEE), gross primary production (GPP), and ecosystem respiration (R<inline-formula><tex-math>$_{text{E}}$</tex-math></inline-formula>) are evaluated against measurements from 37 eddy covariance towers deployed over the ASZ, spanning from 2015 to 2022. The assessment indicates that the NEE unbiased root-mean-square error falls within the targeted accuracy of 1.6 gC.m<inline-formula><tex-math>$^{text{-2}}$</tex-math></inline-formula>.d<inline-formula><tex-math>$^{text{-1}}$</tex-math></inline-formula>, as defined for the SPL4CMDL product. However, modeled GPP and R<inline-formula><tex-math>$_{text{E}}$</tex-math></inline-formula> are overestimated at the beginning of the growing season over evergreen needleleaf forests and shrublands, while being underestimated over grasslands. Discrepancies are also found in the annual net CO<inline-formula><tex-math>$_{2}$</tex-math></inline-formula> budgets. SM appears to have a minimal influence on the GPP and R<inline-formula><tex-math>$_{text{E}}$</tex-math></inline-formula> modeling, suggesting that ASZ vegetation is rarely subjected to hydric stress, which contradicts some recent studies. These results highlight the need for further carbon cycle process understanding and model refinements to improve the SPL4CMDL CO<inline-formula><tex-math>$_{text{2}}$</tex-math></inline-formula> flux estimatesover the ASZ.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"9850-9864"},"PeriodicalIF":4.7,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10945423","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143850912","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Khatri-Rao Factorization Based Bi-Level Support Vector Machine for Hyperspectral Image Classification","authors":"Xiaotao Wang","doi":"10.1109/JSTARS.2025.3556351","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3556351","url":null,"abstract":"As a benchmark supervised learning algorithm, support vector machine (SVM) has drawn much attention and reported plenty of impressive results in hyperspectral images (HSIs) classification. It builds decision plane with soft margins to divide data into different classes. In previous studies, SVM usually employs a prepared spatial feature to improve its classification performance. Unlike those available where feature and classifier are separately designed, in this study, a bi-level joint optimization framework is developed to bridge SVM classifier training with Gabor feature learning. It is called bi-level support vector machine (BSVM). Inside BSVM, two data-oriented schemes are designed for further enhancement. First, it utilizes Khatri-Rao factorization to reshape the feature learning problem into tensor form which intends to break the feature factor matrix into small pieces and make the problem feasible in computation. Second, it embeds a local regularization term to promote discriminant ability. The normal vector of BSVM and feature factor matrices are solved by alternating iteration. BSVM is validated by extensive experiments on four popular HSI data sets. It achieves 76.10%, 82.84%, 89.04%, and 89.83% in classification accuracy on Houston 2018, Xiong'an, Houston 2013, and Indian Pines respectively, showing significantly improvement over the latest deep learning algorithms and proving its effectiveness and superiority.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"9636-9649"},"PeriodicalIF":4.7,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10946163","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143850906","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SCASNet: Spatial Context-Aware Selection Network for Small Object Detection in Aerial Imagery","authors":"Zhenkuan Wang;Xue-Mei Dong;Yongli Xu","doi":"10.1109/JSTARS.2025.3555627","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3555627","url":null,"abstract":"The detection of small objects within intricate backgrounds poses a significant challenge in the domain of aerial image object detection. In this manuscript, a spatial context-aware selection network (SCASNet) is proposed, which innovatively integrates a state space model with the YOLO architecture to address this challenge. A spatial selection block and a context-aware block are designed to form a spatial context-aware selection module, which can overcome the limitations of the original state space model in sequence modeling, such as insufficient receptive fields and weak local dependency modeling. Then, a channel prior multidimensional attention enhancement module is proposed to focus on key information and optimize the extraction of spatial relationships. It leverages multiscale strip convolutions to map spatial relationships and dynamically allocates weights across channel and spatial dimensions. Finally, a content-focused attention module is designed in the detection heads to fuse fine-grained features from the lower layers of the backbone network with semantic features from the neck layers, which enhances the richness of feature representation. Extensive experiments conducted on publicly available datasets, VisDrone, AI-TOD, and SSDD, demonstrate the competitive performance of the proposed SCASNet compared with existing aerial image object detection models.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"9351-9367"},"PeriodicalIF":4.7,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10944567","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143845539","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Investigation of Spectral Variations in Different Solar Photovoltaic Modules Using Spaceborne and In-Situ Hyperspectral Data","authors":"Shoki Shimada;Hiroki Mizuochi;Wataru Takeuchi","doi":"10.1109/JSTARS.2025.3555609","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3555609","url":null,"abstract":"Photovoltaic (PV) technology is critical to achieving a sustainable society. These PV systems are distributed over large areas, making it important to study their characteristics for applications such as solar power output modeling. The combination of satellite imagery, especially multispectral imagery, with machine learning models has proven to be an effective tool for locating PVs. However, PVs are typically treated as a single category in previous studies, despite the existence of different types with different important parameters such as energy conversion efficiency and lifetime. The objective of this research was to investigate the detailed spectral data of four PV types currently on the market to enable their differentiation in hyperspectral satellite data. Spectral samples were collected using a handheld spectrometer and the hyperspectral imager suite satellite hyperspectral sensor. There were notable differences in the reflectance characteristics of the four PV types. The presence of vegetation around the PVs can affect the satellite-encoded spectral signature. Four spectral indices (SIs) were defined based on the spectral characteristics of each PV type, and the discriminability of the different PV types based on the proposed indices was evaluated using a statistical metric. The potential to discriminate between different solar PV types was demonstrated through the combined use of satellite hyperspectral data and SIs.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"9701-9707"},"PeriodicalIF":4.7,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10945371","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143850890","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SChanger: Change Detection From a Semantic Change and Spatial Consistency Perspective","authors":"Ziyu Zhou;Keyan Hu;Yutian Fang;Xiaoping Rui","doi":"10.1109/JSTARS.2025.3555849","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3555849","url":null,"abstract":"Change detection is a key task in Earth observation applications. Recently, deep learning methods have demonstrated strong performance and widespread application. However, change detection faces data scarcity due to the labor-intensive process of accurately aligning remote sensing images of the same area, which limits the performance of deep learning algorithms. To address the data scarcity issue, we develop a fine-tuning strategy called the semantic change network. We initially pretrain the model on single-temporal supervised tasks to acquire prior knowledge of instance feature extraction. The model then employs a shared-weight Siamese architecture and extended temporal fusion module to preserve this prior knowledge and is fine-tuned on change detection tasks. The learned semantics for identifying all instances is changed to focus on identifying only the changes. Meanwhile, we observe that the locations of changes between the two images are spatially identical, a concept we refer to as spatial consistency. We introduce this inductive bias through an attention map that is generated by large-kernel convolutions and applied to the features from both time points. This enhances the modeling of multiscale changes and helps capture underlying relationships in change detection semantics. We develop a binary change detection model utilizing these two strategies. The model is validated against state-of-the-art methods on six datasets, surpassing all benchmark methods and achieving F1 scores of 92.87%, 86.43%, 68.95%, 97.62%, 84.58%, and 93.20% on the LEVIR-CD, LEVIR-CD+, S2Looking, CDD, SYSU-CD, and WHU-CD datasets, respectively.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"10186-10203"},"PeriodicalIF":4.7,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10945386","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143860874","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Class Imbalance in the Automatic Interpretation of Remote Sensing Images: A Review","authors":"Pengdi Chen;Yuanrui Ren;Baoan Zhang;Yuan Zhao","doi":"10.1109/JSTARS.2025.3555567","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3555567","url":null,"abstract":"Class imbalance is a very challenging problem in data science, affecting the development of several application fields. This problem also plagues the automatic interpretation of remote sensing images. Especially in tasks such as classification mapping, object detection, change detection, and scene classification, the classes of training samples required by machine learning exhibit uneven distribution, which seriously affects the accuracy of model training. Our meta-analysis is based on 171 journal papers retrieved and screened from the Web of Science database, covering publication years, highly productive countries, highly cited authors, remote sensing data types, data augmentation methods, and the distribution of the main application fields. The solution to the proposed problem involves three aspects: model innovation and optimization, loss function improvement, and data augmentation. Experiments on benchmark datasets have demonstrated the effectiveness of these methods. In terms of remote sensing task applications, we provide a comprehensive review and analysis of recent research cases on deep learning aimed at addressing the class imbalance problem. Finally, we discuss the synergistic relationship between models, loss functions, and data augmentation, summarize the current challenges in this field, as well as propose several ideas for addressing the class imbalance problem.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"9483-9508"},"PeriodicalIF":4.7,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10945429","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143845540","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"UAVINE-XAI: eXplainable AI-Based Spectral Band Selection for Vineyard Monitroting Using UAV Hyperspectral Data","authors":"Olympia Kourounioti;Anastasios Temenos;Nikos Temenos;Emmanouil Oikonomou;Anastasios Doulamis;Nikolaos Doulamis","doi":"10.1109/JSTARS.2025.3555788","DOIUrl":"https://doi.org/10.1109/JSTARS.2025.3555788","url":null,"abstract":"An efficient spectral band selection trustworthy machine learning (ML) framework for vineyard monitoring from uncrewed aerial vehicle (UAV) hyperspectral data is introduced. The UAV, equipped with Specim AFX-10, is used to capture data beyond the visible spectrum within the 400–1000 nm wavelength range for a total of 224 bands. Popular supervised ML algorithms are utilized for detecting vegetation canopy in vineyards and distinguishing it from existing land uses, namely ground and shadow. Explainable AI results accompany those from ML to identify the most important bands, and understand the contribution of their reflectance levels to the ML models. By doing so, the number of spectral bands is narrowed while maintaining the granularity of the HS data. Experimental results on UAVINE, a publicly available dataset, demonstrate excelling classification performance of random forest (RF) with an overall accuracy of 97.06%, and with precision, recall, and F1-scores following accordingly. With the use of the computationally efficient Tree SHAP algorithm applied on the RF, the bands B106 (677 nm—Red), B186 (897 nm—NIR), B211 (967 nm—NIR), and B39 (498 nm—Green) were identified as the most important ones to the model, enabling better visualization of the vineyard and band-based analysis for each one of the classes existing within the vineyard.","PeriodicalId":13116,"journal":{"name":"IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing","volume":"18 ","pages":"10095-10104"},"PeriodicalIF":4.7,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10945393","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143850845","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}