IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society最新文献_第8页

MUFNet: A Mamba-Based Uncertainty-Aware Fusion Framework for Fine-Grained Satellite Hyperspectral-LiDAR Tree Species Mapping MUFNet：一种基于mamba的精细卫星高光谱-激光雷达树种测绘的不确定性感知融合框架

IF 4.4

IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society Pub Date : 2026-01-01 Epub Date: 2026-04-07 DOI: 10.1109/LGRS.2026.3681844

Xueqin He;Kaijian Xu;Shuzhou Wang;Ping Zhao;Yuqi Zhang;Juanjuan Bi

{"title":"MUFNet: A Mamba-Based Uncertainty-Aware Fusion Framework for Fine-Grained Satellite Hyperspectral-LiDAR Tree Species Mapping","authors":"Xueqin He;Kaijian Xu;Shuzhou Wang;Ping Zhao;Yuqi Zhang;Juanjuan Bi","doi":"10.1109/LGRS.2026.3681844","DOIUrl":"https://doi.org/10.1109/LGRS.2026.3681844","url":null,"abstract":"Satellite hyperspectral imagery (HSI) fused with LiDAR enables large-area, fine-grained tree-species mapping in subtropical forests. However, current satellite HSI-LiDAR fusion remains limited because nontask-oriented spectral reduction may weaken discriminative narrowband and interband cues, while weakly coupled fusion cannot adequately address cross-modal inconsistency in satellite observations. To address these issues, we propose MUFNet, a mamba-based uncertainty-aware fusion framework that integrates multistage cross-modal interactions with uncertainty-guided prediction aggregation under end-to-end task-driven representation learning. We validate MUFNet in a representative subtropical region and benchmark it against six state-of-the-art (SOTA) models. Experimental results demonstrate that MUFNet achieves the best mapping performance, with an overall accuracy (OA) (kappa) of 93.34% (91.17%). Moreover, MUFNet significantly outperforms all competing methods, improving OA (kappa) by 2.85%–11.45% (6.45%–15.49%). In addition, MUFNet exhibits good predictive reliability, achieving an expected calibration error (ECE) of 0.0073, and maintains a favorable performance–efficiency trade-off with only 0.400 M parameters and 0.086 GFLOPs. Ablation studies further confirm the complementarity of the three fusion stages and the contributions of the proposed task-driven representation learning, cross-modal interaction, and uncertainty-aware routing components. This work provides a transferable technical pathway for fine-grained tree species mapping under multisource remote sensing, offering methodological support for biodiversity conservation and forest resource management.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"23 ","pages":"1-5"},"PeriodicalIF":4.4,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148490166","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}

引用次数: 0

PSL-Flow: Physics-Structured Latent Flow Matching for Aerial Visible-to-Infrared Image Translation PSL-Flow：用于航空可见光到红外图像转换的物理结构潜流匹配

IF 4.4

IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society Pub Date : 2026-01-01 Epub Date: 2026-08-24 DOI: 10.1109/LGRS.2026.3726472

Leping Lin;Zibin Zheng;Ning Ouyang

{"title":"PSL-Flow: Physics-Structured Latent Flow Matching for Aerial Visible-to-Infrared Image Translation","authors":"Leping Lin;Zibin Zheng;Ning Ouyang","doi":"10.1109/LGRS.2026.3726472","DOIUrl":"https://doi.org/10.1109/LGRS.2026.3726472","url":null,"abstract":"Aerial visible-to-infrared (V2IR) image translation provides scalable data support for nighttime remote sensing. However, existing methods predominantly learn cross-modal mappings within pixel or generic latent spaces, where physical information acts merely as an auxiliary training constraint. This limitation hinders physical cues from continuously participating in thermal structure generation during inference. In this letter, we propose PSL-Flow, a novel framework for aerial V2IR image generation via conditional flow matching within a physics-structured latent space (PSL-VAE). PSL-Flow shifts the generative objective from direct pixel mapping to physical quantity distribution modeling, enabling physical information to serve as intrinsic generative variables. Specifically, physical quantities including temperature are estimated from real infrared (IR) images to construct an intermediate representation endowed with thermal imaging semantics. Subsequently, the physical backbone and local residuals are jointly encoded into a PSL-VAE, transforming the objective from pixel mapping to thermal state modeling. Ultimately, a continuous generation process driven by visible conditions is learned within this latent space, and IR images are reconstructed through the synergistic decoding of the physical backbone and boundary residuals. Experimental results demonstrate that the proposed method achieves stable advantages across multiple aerial datasets. Our code is available at <uri>https://github.com/zbZ113/PSL-Flow</uri>","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"23 ","pages":"7002505-7002505"},"PeriodicalIF":4.4,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148925493","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}

引用次数: 0

Three-Branch Hybrid Network for Farmland Segmentation in Remote Sensing Images 基于三分支混合网络的遥感影像农田分割

IF 4.4

IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society Pub Date : 2026-01-01 Epub Date: 2026-08-25 DOI: 10.1109/LGRS.2026.3727247

Weihui Zeng;Fang Wang;Gensheng Hu

{"title":"Three-Branch Hybrid Network for Farmland Segmentation in Remote Sensing Images","authors":"Weihui Zeng;Fang Wang;Gensheng Hu","doi":"10.1109/LGRS.2026.3727247","DOIUrl":"https://doi.org/10.1109/LGRS.2026.3727247","url":null,"abstract":"The accurate segmentation of remote sensing imagery is critical for precision agriculture but challenging due to spectral complexity and ambiguous interclass boundaries. The convolutional neural networks are limited in modeling global context, while transformer-based methods incur high computational overhead. This letter proposes a three-branch hybrid network (TBHNet). A collaborative global branch introduces an attention-enhanced visual state-space module (AVSSM) by integrating Mamba-based visual state-space blocks with an attention mechanism to enhance global context modeling. Parallel spatial and boundary branches preserve local structures and refine edges for complementary multibranch feature fusion. A lightweight multireceptive group convolution head (MRHead) is further designed to improve prediction efficiency. Experiments on FGFD and LoveDA datasets show that TBHNet achieves mean intersection over union (mIoU) of 85.90% and 68.58% and overall accuracy (Acc) of 92.69% and 84.46%, respectively, striking a superior balance between segmentation performance and computational cost.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"23 ","pages":"2506405-2506405"},"PeriodicalIF":4.4,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148925847","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}

引用次数: 0

STA-Net: Spatio-Temporal Axial Transformer for Semantic Change Detection in High-Resolution Remote Sensing Imagery STA-Net：用于高分辨率遥感图像语义变化检测的时空轴向变压器

IF 4.4

IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society Pub Date : 2026-01-01 Epub Date: 2026-08-13 DOI: 10.1109/LGRS.2026.3723520

Xin Wang;Ying Gao;Li Li

{"title":"STA-Net: Spatio-Temporal Axial Transformer for Semantic Change Detection in High-Resolution Remote Sensing Imagery","authors":"Xin Wang;Ying Gao;Li Li","doi":"10.1109/LGRS.2026.3723520","DOIUrl":"https://doi.org/10.1109/LGRS.2026.3723520","url":null,"abstract":"Remote sensing (RS) semantic change detection (SCD) confronts key challenges of temporal feature misalignment, fragmented narrow-elongated target prediction, and inconsistent semantic change representations due to temporal variations and intricate geographic target properties. To address these intertwined challenges, this letter proposes STA-Net, a multitask framework integrating three synergistic core modules: cross-temporal interaction (CTI), axial-semantic transformer (AST), and spatio-temporal fusion (STF). The CTI module aligns cross-temporal features via channel subset swapping to mitigate misalignment, the AST module preserves narrow-elongated target structural integrity through axial attention (AA) for fine-grained semantic representation, and the STF module fuses spatio-temporal cues via 3-D convolution to resolve semantic change inconsistency, forming a closed-loop “alignment-representation-fusion” collaborative framework that addresses each bottleneck. On the SECOND and HRSCD datasets, STA-Net achieves state-of-the-art fine-grained SCD performance and superior robustness across multiple metrics [e.g., FSCD and weighted score (WS)]. With a multitask architecture for high-resolution RS scenes, it delivers high-precision detection with a favorable tradeoff between computational efficiency and detection performance, demonstrating strong effectiveness for RS-SCD tasks. The source code is available at: <uri>https://github.com/WangXin81/STA-Net/</uri>","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"23 ","pages":"6017805-6017805"},"PeriodicalIF":4.4,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148925922","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}

引用次数: 0

Coupled Geometry-Sequence State-Space Model for Point Cloud Semantic Segmentation 点云语义分割的耦合几何-序列状态空间模型

IF 4.4

IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society Pub Date : 2026-01-01 Epub Date: 2026-08-25 DOI: 10.1109/LGRS.2026.3726965

Dongrong Yang;Siyuan Hao;Yuanxin Ye

{"title":"Coupled Geometry-Sequence State-Space Model for Point Cloud Semantic Segmentation","authors":"Dongrong Yang;Siyuan Hao;Yuanxin Ye","doi":"10.1109/LGRS.2026.3726965","DOIUrl":"https://doi.org/10.1109/LGRS.2026.3726965","url":null,"abstract":"The complex spatial structures and varying scales of point cloud data make semantic segmentation a highly challenging task. While downsampling and feature fusion are essential steps in mainstream networks, existing models often suffer from local geometry loss and multiscale semantic conflicts during these processes. To alleviate these issues, we propose coupled geometry-sequence (CGS)-Mamba, a novel state-space model (SSM)-based network for point cloud semantic segmentation. Specifically, we propose the CSG block (CGS-Block). It extracts fine-grained local geometric features and effectively captures global long-range dependencies. Furthermore, we introduce the adaptive gated fusion (AGF) module in the decoder, which can mitigate semantic conflicts during cross-layer feature fusion through on-demand feature absorption. Experiments show that CGS-Mamba achieves mean intersection over union (mIoU) scores of 86.83% and 75.46% on the Tinto and Toronto-3-D datasets, respectively, along with overall accuracy (OA) of 94.59% and 93.35%. Notably, compared to baselines, our model achieves significant mIoU improvements of 1.53% and 2.53% on the two datasets, respectively, while nearly halving the parameters to 16.16 M. The code and models are available at <uri>https://github.com/MMYY-LL/CGS-Mamba</uri>","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"23 ","pages":"6501205-6501205"},"PeriodicalIF":4.4,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148925937","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}

引用次数: 0

Semantic-Guided Fusion Network for Multisource Remote Sensing Image Classification 多源遥感图像分类的语义引导融合网络

IF 4.4

IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society Pub Date : 2026-01-01 Epub Date: 2026-08-24 DOI: 10.1109/LGRS.2026.3726463

Yuwei Zhao;Chuanzheng Gong;Baogui Huan;Feng Gao;Junyu Dong;Qian Du

{"title":"Semantic-Guided Fusion Network for Multisource Remote Sensing Image Classification","authors":"Yuwei Zhao;Chuanzheng Gong;Baogui Huan;Feng Gao;Junyu Dong;Qian Du","doi":"10.1109/LGRS.2026.3726463","DOIUrl":"https://doi.org/10.1109/LGRS.2026.3726463","url":null,"abstract":"Multisource remote sensing image classification has attracted increasing attention due to the complementary spectral, structural, and geometric information. However, existing methods still suffer from two limitations: insufficient semantic contextual modeling and unreliable feature fusion caused by slight spatial misalignment. To address these issues, we propose a semantic-guided fusion network (SGFNet) for multisource remote sensing image classification. Specifically, the semantic mixing convolution block (SMCB) is designed to dynamically generate semantic-aware convolution kernels according to contextual relationships among feature representations. In addition, the frequency-modulated fusion block (FMFB) is introduced to perform cross-modal interaction in the frequency domain, which effectively alleviates the influence of slight spatial misalignment and improves complementary information fusion. Extensive experiments conducted on the Augsburg and Houston 2018 datasets demonstrate that the proposed SGFNet consistently outperforms several state-of-the-art methods. The codes are publicly available at <uri>https://github.com/oucailab/SGFNet</uri>","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"23 ","pages":"5507305-5507305"},"PeriodicalIF":4.4,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148925888","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}

引用次数: 0

Coupling Ray Tracing, SNICAR, and DISORT to Facilitate Snow Albedo and Bidirectional Reflectance Modeling 耦合光线追踪、SNICAR和DISORT促进雪反照率和双向反射建模

IF 4.4

IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society Pub Date : 2026-01-01 Epub Date: 2026-08-24 DOI: 10.1109/LGRS.2026.3726465

Gongxue Wang;Qi Wang;Lingmei Jiang;Jie Shao;Donghui Xie;Guangjin Liu

{"title":"Coupling Ray Tracing, SNICAR, and DISORT to Facilitate Snow Albedo and Bidirectional Reflectance Modeling","authors":"Gongxue Wang;Qi Wang;Lingmei Jiang;Jie Shao;Donghui Xie;Guangjin Liu","doi":"10.1109/LGRS.2026.3726465","DOIUrl":"https://doi.org/10.1109/LGRS.2026.3726465","url":null,"abstract":"Spectral albedo and anisotropy of the snow surface can be substantially modulated by variations in the layers’ microstructure and composition of snow particles. A persistent gap exists between models that consider optimal snow microstructure and composition and practical snow optics modeling. We propose an advanced snow optics model framework that incorporates the complex effects of snow grain shape and size, the deposition of various light-absorbing constituents (LACs), and their vertical profiles, thereby enabling accurate joint modeling of albedo and bidirectional reflectance. The model is established by coupling the single-scattering properties (SSPs) from ray-tracing simulations and the snow, ice, and aerosol radiative (SNICAR) model with the discrete ordinates radiative transfer (DISORT) solver. Taking the Mishchenko method as a reference, 16-stream DISORT yields more accurate albedo than the two-stream approximation of SNICAR, which becomes significant for incident angles larger than 60°. Compared against in situ snow measurements, the proposed model exhibits promising accuracy with root-mean-square error (RMSE) <inline-formula> <tex-math>$leq$ </tex-math></inline-formula>0.1 for bidirectional reflectance. The model’s flexible architecture also enables the future integration of more comprehensive representations of snow and ice properties.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"23 ","pages":"2000605-2000605"},"PeriodicalIF":4.4,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148926106","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}

引用次数: 0

A Coarse-to-Fine Progressive Fusion Network for Multimodal Semantic Segmentation of Remote Sensing Images 一种用于遥感图像多模态语义分割的粗到精渐进融合网络

IF 4.4

IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society Pub Date : 2026-01-01 Epub Date: 2026-07-31 DOI: 10.1109/LGRS.2026.3718592

Di Zhang;Yuhang Yan;Qianqian Niu;Peicheng Yue

{"title":"A Coarse-to-Fine Progressive Fusion Network for Multimodal Semantic Segmentation of Remote Sensing Images","authors":"Di Zhang;Yuhang Yan;Qianqian Niu;Peicheng Yue","doi":"10.1109/LGRS.2026.3718592","DOIUrl":"https://doi.org/10.1109/LGRS.2026.3718592","url":null,"abstract":"Multimodal semantic segmentation of high-resolution remote sensing imagery is important for fine-grained land-cover interpretation. However, existing fusion methods still suffer from unstable shallow optical-DSM alignment and deep feature degradation caused by heterogeneous frequency noise, boundary-detail loss, and inconsistent spatial responses. To address the aforementioned challenges, this letter proposes a coarse-to-fine progressive fusion network (CFPFNet). Specifically, a visual state space model extracts a Mamba-derived global structural prior to guide the coarse-grained context enhancement (CGCE) module for preliminary cross-modal alignment. Then, the fine-grained adaptive frequency-spatial fusion (FGAF) module performs amplitude-phase collaboration and adaptive spatial cross-gating for multiscale semantic refinement. Experiments on the ISPRS Vaihingen and Potsdam datasets demonstrate the effectiveness of CFPFNet. On Vaihingen, CFPFNet improves mIoU and mF1 by 2.22% and 1.39% over the simple dual-stream baseline, respectively.","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"23 ","pages":"7506705-7506705"},"PeriodicalIF":4.4,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148780909","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}

引用次数: 0

Low-Rank Spectral Continuity Modeling With Gaussian Splatting for Hyperspectral Image Compression 基于高斯溅射的高光谱图像压缩低秩光谱连续性建模

IF 4.4

IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society Pub Date : 2026-01-01 Epub Date: 2026-06-30 DOI: 10.1109/LGRS.2026.3708431

Lili Zhang;Yukun Li;Yang Gao;Jiahui Liu

{"title":"Low-Rank Spectral Continuity Modeling With Gaussian Splatting for Hyperspectral Image Compression","authors":"Lili Zhang;Yukun Li;Yang Gao;Jiahui Liu","doi":"10.1109/LGRS.2026.3708431","DOIUrl":"https://doi.org/10.1109/LGRS.2026.3708431","url":null,"abstract":"Lossy compression has become the mainstream approach for downlinking hyperspectral images (HSIs). However, most learning-based methods rely on implicit mappings that fail to preserve inherent structural priors of HSIs, while existing explicit methods lack end-to-end optimization and spectral continuity modeling. To address these issues, a low-rank spectral continuity modeling with a Gaussian splatting framework (LR-GS) for HSI compression is proposed. Singular value decomposition (SVD) and rank-revealing QR (RRQR) construct a compact low-rank spectral representation with band selection and stable spectral basis vector estimation. Representation coefficients are modeled as continuous functions using weighted Gaussian primitives. A 3-D total variation (3DTV) constraint is further introduced in the rate–distortion optimization framework to regularize spatial–spectral gradients and suppress artifacts. Experiments show superior performance at different bitrates. The proposed method achieves 36 dB PSNR, improving results by 22% over JPEG2000 and 5% over LoR-SGS at 0.2 bits per pixel per band (bpppb).","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"23 ","pages":"5507705-5507705"},"PeriodicalIF":4.4,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148986121","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}

引用次数: 0

Infrared Small Target Detection via Spatio-Temporal Manifold Prior and Structural Rectification 基于时空流形先验和结构校正的红外小目标检测

IF 4.4

IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society Pub Date : 2026-01-01 Epub Date: 2026-09-07 DOI: 10.1109/LGRS.2026.3731139

Anqing Wu;Qiang Guo;Mykola Kaliuzhnyi

{"title":"Infrared Small Target Detection via Spatio-Temporal Manifold Prior and Structural Rectification","authors":"Anqing Wu;Qiang Guo;Mykola Kaliuzhnyi","doi":"10.1109/LGRS.2026.3731139","DOIUrl":"https://doi.org/10.1109/LGRS.2026.3731139","url":null,"abstract":"Infrared small target detection in dynamic scenes is challenging because complex background structures are not globally low-rank. Directly imposing low-rank regularization on the raw background can over-smooth high-frequency content and transfer structural residuals into the sparse-target component. To address this deficiency, we propose a spatio-temporal manifold prior and structural rectification method. First, to bypass rigid global assumptions, we construct a spatio-temporal patch matrix and employ a KD-tree-accelerated local linear embedding framework. This extracts local geometric correlations to establish a data-driven manifold prior, enabling linear prediction of complex backgrounds without annotated training data. Second, recognizing that manifold approximation errors in dynamic scenes manifest as correlated structural artifacts rather than random noise, we model them as a low-dimensional subspace. By shifting the nuclear-norm constraint from the raw background to the manifold residual, structural rectification absorbs these artifacts before sparse-target recovery. Finally, an adaptive ADMM solver is formulated to optimize the proposed model. Experiments on five scenes from two public datasets demonstrate improved background suppression and target preservation in complex dynamic backgrounds. The source code is publicly available at <uri>https://github.com/w-a-q/MPSR.git</uri>","PeriodicalId":91017,"journal":{"name":"IEEE geoscience and remote sensing letters : a publication of the IEEE Geoscience and Remote Sensing Society","volume":"23 ","pages":"7003005-7003005"},"PeriodicalIF":4.4,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149033480","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}

引用次数: 0