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

{"title":"Low-Cost MLS-Based Forest Plot Mapping via Feature Graph Registration","authors":"Qin Ye;Yujia Jin;Junqi Luo","doi":"10.1109/LGRS.2025.3589100","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3589100","url":null,"abstract":"Forest plot mapping is a significant task in forest inventories by providing accurate structural parameters. However, understory mapping still predominantly relies on terrestrial laser scanning (TLS), which is time-consuming and labor-intensive. Moreover, existing mobile laser scanning (MLS)-based methods either require expensive high-beam LiDAR or struggle with feature extraction and registration accuracy. To address these issues, we propose a novel low-cost MLS-based forest plot mapping method utilizing feature graph registration. Local submaps are first constructed via tree stem extraction and scan-to-scan graph-based registration, followed by global alignment to generate the final forest plot map. Experiments on three forest plots with varying structures and species demonstrate that our method achieves an average mapping accuracy of approximately 10 cm, even without loop closure optimization. Comparative results further demonstrate our effectiveness and efficiency for practical forest surveys.","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":0.0,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144671244","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":"A Lightweight Hybrid Network for Object Detection in Remote Sensing Images Balancing Global and Local Information","authors":"Shuting Huang;Ge Zhang;Huanzun Zhang;Hui Xu;Guangzhen Yao;Sandong Zhu;Long Zhang;Jun Kong","doi":"10.1109/LGRS.2025.3588788","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3588788","url":null,"abstract":"In recent years, hybrid convolutional neural networks (CNNs) and Transformer-based object detection technologies have achieved remarkable success. In the field of remote sensing image detection, since remote sensing systems rely on the large-scale deployment of edge devices, detection models need to be lightweight with low parameter complexity to adapt to resource-constrained environments. However, existing lightweight models often struggle with an imbalance in extracting low-frequency global and high-frequency local information. In particular, when processing high-frequency local information (such as edges, textures, and fine structures), these models often lack in-depth analysis, leading to insufficient extraction of local features and reduced detection accuracy. To address the imbalance between low-frequency global information and high-frequency local information in lightweight remote sensing models, we propose an efficient and lightweight hybrid network detection framework, which mainly consists of the global–local balance (GLB) module and the detail-aware feature fusion (DAFF) module. The GLB module adopts dynamic weight adjustment and context-aware mechanisms to effectively aggregate high-frequency local information in the image. The DAFF module further enhances feature fusion and detail refinement, improving the model’s performance and generalization ability. Experimental results on remote sensing datasets, including RSOD, NWPU VHR-10, and LEVIR datasets, demonstrate that our proposed method achieves a well-balanced tradeoff between model size and detection accuracy, reaching state-of-the-art performance.","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":0.0,"publicationDate":"2025-07-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144695550","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":"Collaborative Representation-Based Attention Network for Hyperspectral Anomaly Detection","authors":"Maryam Imani;Daniele Cerra","doi":"10.1109/LGRS.2025.3588163","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3588163","url":null,"abstract":"The collaborative representation-based detector (CRD) performs anomaly detection for hyperspectral (HS) data using a linear representation of local neighbors for background estimation, which may not fully capture the informational content and spectral variability in complex HS images with heterogenous background. To deal with this aspect, the collaborative representation-based attention network (CRAN) is introduced in this letter, providing a nonlinear representation of data samples for background estimation. Both local neighbors and global samples are used in parallel, and their outputs are fused through a cross-attention mechanism. Experimental results show a good performance of CRAN in comparison with several state-of-the-art anomaly detectors.","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":0.0,"publicationDate":"2025-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144705180","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":"On the Detection of the Terra Nova Bay Open Polynya Dynamic Phases Using a Single C-Band SAR Image","authors":"Giovanna Inserra;Ferdinando Nunziata;Andrea Buono;Giuseppe Aulicino;Maurizio Migliaccio","doi":"10.1109/LGRS.2025.3588103","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3588103","url":null,"abstract":"In this study, a novel approach is proposed to extract dynamic information about Terra Nova Bay (TNB) polynya, Antarctica, using a single synthetic aperture radar (SAR) imagery collected by the Sentinel-1 mission. The proposed approach is based on a joint scattering/morphological and spectral analysis of key features, namely, the sea ice streaks, which characterize polynyas. A set of single- and dual-polarimetric Sentinel-1 SAR measurements collected over the open TNB polynya under different growing and closing phases is used for the experimental analysis, which is assisted by ancillary optical and radiometer satellite products. Experimental results demonstrate the ability of copolarized backscattering to identify the open polynya growing and closing phases using a single SAR scene. This suggests using SAR measurements to fill the temporal and spatial gaps arising from the monitoring of the harsh TNB polynya environment.","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":0.0,"publicationDate":"2025-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144705047","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":"Calibration of a Radar Polarimetric Decomposition Using a Radiative Transfer Model","authors":"Giovanni Anconitano;Lorenzo Giuliano Papale;Leila Guerriero;Mario Alberto Acuña;Nazzareno Pierdicca","doi":"10.1109/LGRS.2025.3588254","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3588254","url":null,"abstract":"This letter describes a procedure based on the radiative transfer theory to calibrate the scattering contributions from the Generalized Freeman–Durden (GFD) polarimetric decomposition over corn fields. The Tor Vergata electromagnetic model (TOV) is used to simulate canonical scattering mechanisms that are compared with those obtained by applying GFD to both simulated and L-band SAOCOM-1A data. The proposed method first analyzes the error between the model and the GFD applied to the simulated data. A multivariate data fitting is then performed to derive a new expression of the GFD powers, which is tested on the L-band real data. The GFD volume power obtains the greatest benefit from the calibration, reducing the root mean square error (RMSE) with respect to the corresponding TOV model contribution to 0.006 in linear units. To further test the procedure, a linear regression model is used to estimate soil moisture using the calibrated GFD powers from SAOCOM-1A real data. The retrieval performance, evaluated through a Leave-One-Out (LOO) cross-validation against in-situ data, shows a significant improvement. The calibrated GFD powers lead to an increased linear correlation (from 0.32 to 0.57), while the RMSE is reduced (from 0.096 to 0.055 m³/m³).","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":0.0,"publicationDate":"2025-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11078365","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144687809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Snowpack Permittivity Retrieval Using Particle Swarm Optimization Algorithm","authors":"Lekhmissi Harkati","doi":"10.1109/LGRS.2025.3587507","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3587507","url":null,"abstract":"This letter deals with the estimation of the refractive index of snowpack based on particle swarm optimization (PSO) algorithm. By leveraging its ability to handle multidimensional optimization problems, PSO is used both to compute the shortest path between the synthetic aperture radar tomographic system antenna and the imaged snowpack scatterers, taking into account its permittivity and to optimize the difference in depth coordinates between the snowpack tomogram and a simulated one using the same system configuration and measurement parameters. The proposed method is applied on two measured tomogram, namely, those whose data were acquired during a measurement campaign carried out in the French Alps.","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":0.0,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144663723","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 Hyperspectral Band Selection Using GA-SR-NMI-VI: A Hybrid Similarity and Evolutionary-Based Approach","authors":"Neeraj Kumar Nadipelli;T. Hitendra Sarma;R. Dharma Reddy;Kovvur Ram Mohan Rao;K. Mrudula;Murali Kanthi","doi":"10.1109/LGRS.2025.3587604","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3587604","url":null,"abstract":"Hyperspectral image (HSI) analysis requires effective band selection techniques to enhance classification accuracy while maintaining computational efficiency. Similarity-based ranking normalized mutual information and variation of information (SR-NMI-VI) is a recent SR approach that leverages NMI and VI to compute band rankings. This letter presents an extended version of SR-NMI-VI called genetic algorithm (GA)-SR-NMI-VI, an advanced feature selection approach that integrates GA with similarity-based ranking. The proposed GA-SR-NMI-VI is a two-step approach, where SR-NMI-VI is first used for band ranking, followed by a GA-based optimization to determine the optimal number of bands. For comparative evaluation, two additional methods are considered: GA-SR-structural similarity index (SSIM)-NMI-VI, which adds SSIM to enhance spatial–spectral ranking, and graph-regularized fast and robust principal component analysis (GR-FRPCA), an unsupervised low-rank clustering-based approach. To validate the proposed approaches, an experimental study has been conducted on various HSI datasets covering a diverse range of classes from 2 to 16 and covering different classification scenarios, including Oil Spill, Cubert Drone, WHU-Hi-LongKou, and WHU-Hi-HanChuan. Empirically, it is shown that the proposed GA-SR-NMI-VI and its SSIM-enhanced variant consistently achieve higher accuracy and kappa scores across various machine learning models. GA-SR-NMI-VI achieves a 70%–80% reduction in the number of bands while improving classification accuracy by 2%–5%. Notably, traditional classifiers such as random forest and support vector machine (SVM) perform comparably to deep learning models while benefiting from lower computational costs, highlighting the effectiveness of GA-based methods in scenarios where deep learning may be computationally expensive or infeasible. The details of the experimental setup and the reproducible code are available at the following link: <uri>https://github.com/neerajkumarnadipelli/GA-SR-NMI-VI</uri>","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":0.0,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144657371","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":"Determining Terrestrial Ecosystem Gross Primary Productivity From PACE OCI","authors":"Karl Fred Huemmrich;Skye Caplan;John A. Gamon;Petya Krasteva Entcheva Campbell","doi":"10.1109/LGRS.2025.3587584","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3587584","url":null,"abstract":"Data from the Plankton, Aerosol, Cloud, ocean Ecosystem (PACE) Ocean Color Instrument (OCI) were used to develop and test algorithms for remotely retrieving terrestrial ecosystem productivity. Gross primary productivity (GPP) was calculated from CO2 flux for 47 eddy covariance flux towers representing vegetation and climatic variability across the USA. Eight-day average GPP was matched with eight-day average mapped OCI reflectance data containing 49 spectral bands from ultraviolet through short wave infrared spectral regions. The data covered the growing season from March through September 2024. For the combination of all sites and dates, the red-edge chlorophyll index alone described 66% of the variation in GPP. Using a partial least squares regression (PLSR) on all spectral bands GPP retrieval was improved to 74%. Agricultural sites were often found to have high residuals in these regressions. By training PLSR by eco-climatic region, the overall GPP retrievals were improved to 86%. The success of these algorithms across multiple sites with different vegetation types and through the growing season demonstrates the utility of PACE OCI data to map GPP dynamics at continental scales.","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":0.0,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11075694","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144680871","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hierarchical Multiscale Representation in Remote Sensing Scene Classification","authors":"João Pedro O. Batisteli;Silvio Jamil F. Guimarães;Zenilton K. G. Patrocínio","doi":"10.1109/LGRS.2025.3587580","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3587580","url":null,"abstract":"Remote sensing scene classification (RSSC) poses significant challenges due to high spatial variability, complex textures, and semantic ambiguity in remote sensing imagery. While convolutional neural networks (CNNs) and transformer-based models have achieved notable success in this domain, their performance often depends on large-scale pretraining and substantial computational resources. Graph neural networks (GNNs) have emerged as a promising alternative to traditional deep learning methods by explicitly modeling the relational structure of image regions through graph representations, which have already demonstrated promising results across various image-based tasks involving images. In this work, we explore two GNN architectures tailored for RSSC: BRMv2, a novel simplified graph model built on a base region adjacency graph (RAG), and modified hierarchical layered multigraph network (mHELMNet), a modified hierarchical multigraph model that encodes multiscale and spatial relationships through a multigraph representation. Both models were evaluated on the EUROSAT and RESISC45 datasets, achieving accuracy comparable to, or in some cases exceeding, that of state-of-the-art CNN-based, hybrid GNN-based, and transformer-based methods, while using significantly fewer parameters and without relying on pretraining. Experimental results demonstrated that the proposed GNN models, mHELMNet and BRMv2, achieved over 96% accuracy on EUROSAT and approximately 85% on RESISC45, while requiring only 0.14% and 0.03% of the parameters of the leading transformer-based approach, respectively.","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":0.0,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144680762","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":"Sparse Deformable Mamba for Hyperspectral Image Classification","authors":"Lincoln Linlin Xu;Yimin Zhu;Zack Dewis;Zhengsen Xu;Motasem Alkayid;Mabel Heffring;Saeid Taleghanidoozdoozan","doi":"10.1109/LGRS.2025.3587256","DOIUrl":"https://doi.org/10.1109/LGRS.2025.3587256","url":null,"abstract":"Although Mamba models significantly improve hyperspectral image (HSI) classification, one critical challenge is the difficulty in building the sequence of Mamba tokens efficiently. This letter presents a sparse deformable Mamba (SDMamba) approach for enhanced HSI classification, with the following contributions. First, to enhance the Mamba sequence, an efficient sparse deformable sequencing (SDS) approach is designed to adaptively learn the “optimal” sequence, leading to a sparse and deformable Mamba sequence with increased detail preservation and decreased computations. Second, to boost spatial–spectral feature learning, based on SDS, a sparse deformable spatial Mamba module (SDSpaM) and a sparse deformable spectral Mamba module (SDSpeM) are designed for tailored modeling of the spatial information spectral information. Last, to improve the fusion of SDSpaM and SDSpeM, an attention-based feature fusion approach is designed to integrate the outputs of the SDSpaM and SDSpeM. The proposed method is tested on three benchmark datasets with many state-of-the-art approaches, including convolutional neural networks (CNNs), GAN Transformer, and Mamba-based methods, demonstrating that the proposed approach can achieve higher accuracy with less computation, and better detail small-class preservation capability.","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":0.0,"publicationDate":"2025-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144680842","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}