{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,22]],"date-time":"2026-10-22T16:20:06Z","timestamp":1792686006898,"version":"4.5.0"},"reference-count":51,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,4,23]],"date-time":"2026-04-23T00:00:00Z","timestamp":1776902400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurorobot."],"abstract":"<jats:p>Deep learning techniques have been successfully applied to object classification in Synthetic Aperture Radar (SAR) images, achieving remarkable performance. However, the current Transformer architecture still faces two limitations in SAR object classification: (1) Transformers occupies substantial GPU memory during training, which hinders large-scale model deployment and hyperparameter optimization. (2) the vectorization of image patches destroys spatial information among pixels, thus the represented features are semantic-agnostic. In this paper, we propose a lightweight and interpretable vision mamba (IVim) whose feature maps from deep Mamba blocks are visually understandable. IVim consists of two modules: Token Semantic Re-allocation (TSR) and Token Attention Selection (TSA). TSR re-allocates the semantics to feature maps by optimizing a novel loss function for an extra convolutional layer. TAS selects the most discriminative attention map by evaluating the quantity of semantics in attention matrices of different Mamba layer. Finally, we provide a coupling strategy to form a saliency heatmap to visually show the interpretability of IVim by merging feature map and attention map. Experimental results demonstrate that IVim can re-allocate semantics to deep features and achieve better classification performance compared to its counterparts by reducing 67.8% parameters.<\/jats:p>","DOI":"10.3389\/fnbot.2026.1753927","type":"journal-article","created":{"date-parts":[[2026,4,23]],"date-time":"2026-04-23T05:36:40Z","timestamp":1776922600000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["Interpretable vision mamba for SAR images classification"],"prefix":"10.3389","volume":"20","author":[{"given":"Kaiming","family":"Zou","sequence":"first","affiliation":[{"name":"School of Geophysics, Chengdu University of Technology","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ningbo","family":"He","sequence":"additional","affiliation":[{"name":"School of Geophysics, Chengdu University of Technology","place":["Chengdu, China"]},{"name":"China Aneng Group Third Engineering Bureau Co., Ltd.","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Leijun","family":"Yao","sequence":"additional","affiliation":[{"name":"China Aneng Group Third Engineering Bureau Co., Ltd.","place":["Chengdu, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jialin","family":"Li","sequence":"additional","affiliation":[{"name":"Xi'an Jiuzhi-Tech. Co.","place":["Ltd., China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yongfeng","family":"Wang","sequence":"additional","affiliation":[{"name":"Xi'an Jiuzhi-Tech. Co.","place":["Ltd., China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,4,23]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"17654","DOI":"10.1109\/JSTARS.2025.3587701","article-title":"Rsciwanet: Regional spatial-channel information weighted attention network for video sar and large-scale sar image targets detection","volume":"18","author":"Chang","year":"2025","journal-title":"IEEE J. Select. Topics Appl. Earth Observat. Remote Sens"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2024.3407574","article-title":"A nonparametric estimator for coherent change detection: The permutational change detection","volume":"62","author":"Costa","year":"2024","journal-title":"IEEE Trans. Geosci. Remote Sens"},{"key":"B3","doi-asserted-by":"publisher","first-page":"10065","DOI":"10.1109\/TAES.2025.3560256","article-title":"Multi-view visual and topological features coordination aggregation framework for sar target recognition","volume":"61","author":"Dai","year":"2025","journal-title":"IEEE Trans. Aerospace Electronic Syst"},{"key":"B4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2021.3105575","article-title":"Saenet: Self-supervised adversarial and equivariant network for weakly supervised object detection in remote sensing images","volume":"60","author":"Feng","year":"2022","journal-title":"IEEE Trans. Geosci. Remote Sens"},{"key":"B5","doi-asserted-by":"publisher","first-page":"104","DOI":"10.1016\/j.neucom.2023.02.057","article-title":"Vs-cam: Vertex semantic class activation mapping to interpret vision graph neural network","volume":"533","author":"Feng","year":"2023","journal-title":"Neurocomputing"},{"key":"B6","doi-asserted-by":"publisher","first-page":"106473","DOI":"10.1016\/j.neunet.2024.106473","article-title":"Cluster-cam: Cluster-weighted visual interpretation of cnns decision in image classification","volume":"178","author":"Feng","year":"2024","journal-title":"Neural Networks"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2212.14052","article-title":"Hungry hungry hippos: towards language modeling with state space models","author":"Fu","year":"2022","journal-title":"arXiv"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2312.00752","article-title":"Mamba: Linear-time sequence modeling with selective state spaces","author":"Gu","year":"2023","journal-title":"arXiv"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2111.00396","article-title":"Efficiently modeling long sequences with structured state spaces","author":"Gu","year":"2021","journal-title":"arXiv"},{"key":"B10","first-page":"7226","article-title":"\u201cSar despeckling via regional denoising diffusion probabilistic model,\u201d","author":"Hu","year":"2024","journal-title":"IGARSS 2024"},{"key":"B11","first-page":"87","article-title":"\u201cLong movie clip classification with state-space video models,\u201d","volume-title":"European Conference on Computer Vision","author":"Islam","year":"2022"},{"key":"B12","first-page":"18749","article-title":"\u201cEfficient movie scene detection using state-space transformers,\u201d","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Islam","year":"2023"},{"key":"B13","first-page":"770","article-title":"\u201cDeep residual learning for image recognition,\u201d","volume-title":"2023 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Jing Nathan","year":"2016"},{"key":"B14","doi-asserted-by":"crossref","first-page":"473","DOI":"10.1109\/ICCECT57938.2023.10140842","article-title":"\u201cEffectiveness evaluation of cooperative reconnaissance between electronic signal reconnaissance equipment and sar,\u201d","volume-title":"2023 IEEE International Conference on Control, Electronics and Computer Technology (ICCECT)","author":"Jinke","year":"2023"},{"key":"B15","first-page":"3524","article-title":"\u201cHide-and-seek: Forcing a network to be meticulous for weakly-supervised object and action localization,\u201d","volume-title":"Proceedings of the IEEE International Conference on Computer Vision (ICCV)","author":"Kumar Singh","year":"2017"},{"key":"B16","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/LGRS.2025.3541895","article-title":"MBSSNet: A mamba-based joint semantic segmentation network for optical and sar images","volume":"22","author":"Li","year":"2025","journal-title":"IEEE Geosci. Remote Sens. Letters"},{"key":"B17","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2021.3100137","article-title":"Multiscale CNN based on component analysis for sar atr","volume":"60","author":"Li","year":"2022","journal-title":"IEEE Trans. Geosci. Remote Sens"},{"key":"B18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/LGRS.2022.3212078","article-title":"Self-supervised sar despeckling powered by implicit deep denoiser prior","volume":"19","author":"Lin","year":"2022","journal-title":"IEEE Geosci. Remote Sens. Letters"},{"key":"B19","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2024.3354999","article-title":"Itnet: Low-shot instance transformation network for weakly supervised object detection in remote sensing images","volume":"62","author":"Liu","year":"2024","journal-title":"IEEE Trans. Geosci. Remote Sens"},{"key":"B20","doi-asserted-by":"publisher","first-page":"103031","DOI":"10.52202\/079017-3273","article-title":"VMamba: Visual state space model","volume":"37","author":"Liu","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2401.04722","article-title":"U-Mamba: Enhancing long-range dependency for biomedical image segmentation","author":"Ma","year":"2024","journal-title":"arXiv"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2206.13947","article-title":"Long range language modeling via gated state spaces","author":"Mehta","year":"2022","journal-title":"arXiv"},{"key":"B23","doi-asserted-by":"publisher","first-page":"42190","DOI":"10.1109\/ACCESS.2019.2906564","article-title":"A gradually distilled CNN for sar target recognition","volume":"7","author":"Min","year":"2019","journal-title":"IEEE Access"},{"key":"B24","doi-asserted-by":"publisher","first-page":"2846","DOI":"10.52202\/068431-0206","article-title":"S4nd: Modeling images and videos as multidimensional signals with state spaces","volume":"35","author":"Nguyen","year":"2022","journal-title":"Adv. Neural Inf. Process. Syst"},{"key":"B25","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2025.3566088","article-title":"M3-cr: Multiscale multibranch mamba for sar-assisted optical image thick cloud removal","volume":"63","author":"Pan","year":"2025","journal-title":"IEEE Trans. Geosci. Remote Sens"},{"key":"B26","first-page":"983","article-title":"\u201cAblation-CAM: Visual explanations for deep convolutional network via gradient-free localization,\u201d","volume-title":"Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision (WACV)","author":"Ramaswamy","year":"2020"},{"key":"B27","first-page":"618","article-title":"\u201cGrad-CAM: Visual explanations from deep networks via gradient-based localization,\u201d","volume-title":"Proceedings of the IEEE International Conference on Computer Vision (ICCV)","author":"Selvaraju","year":"2017"},{"key":"B28","first-page":"9640","article-title":"\u201cShip ATR in high resolution SAR images via convolutional transformer,\u201d","author":"Shang","year":"2024","journal-title":"IGARSS 2024"},{"key":"B29","doi-asserted-by":"crossref","first-page":"327","DOI":"10.1109\/ICSIP55141.2022.9887035","article-title":"\u201cMulti-channel sar moving target detection by integrating stap and faster R-CNN,\u201d","volume-title":"2022 7th International Conference on Signal and Image Processing (ICSIP)","author":"Sun","year":"2022"},{"key":"B30","article-title":"\u201cRethinking model scaling for convolutional neural networks,\u201d","volume-title":"Proceedings of the International Conference on Machine Learning","author":"Tan","year":"2019"},{"key":"B31","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/LGRS.2025.3595843","article-title":"Mamba-UDA: Mamba unsupervised domain adaptation for SAR ship detection","volume":"22","author":"Tu","year":"2025","journal-title":"IEEE Geosci. Remote Sens. Letters"},{"key":"B32","author":"Vaswani","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B33","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/LGRS.2022.3183467","article-title":"Global in local: A convolutional transformer for sar atr fsl","volume":"19","author":"Wang","year":"2022","journal-title":"IEEE Geosci. Remote Sens. Letters"},{"key":"B34","first-page":"24","article-title":"\u201cScore-CAM: Score-weighted visual explanations for convolutional neural networks,\u201d","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPR)","author":"Wang","year":"2020"},{"key":"B35","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2023.3295797","article-title":"Remote-sensing scene classification via multistage self-guided separation network","volume":"61","author":"Wang","year":"","journal-title":"IEEE Trans. Geosci. Remote Sens"},{"key":"B36","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-emnlp.5","article-title":"Pretraining without attention","author":"Wang","year":"2022","journal-title":"arXiv"},{"key":"B37","first-page":"6387","article-title":"\u201cSelective structured state-spaces for long-form video understanding,\u201d","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition","author":"Wang","year":""},{"key":"B38","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2021.3074630","article-title":"Combining optimized sar-sift features and rd model for multisource sar image registration","volume":"60","author":"Wang","year":"2022","journal-title":"IEEE Trans. Geosci. Remote Sens"},{"key":"B39","first-page":"1","article-title":"\u201cTarget detection and recognition based on convolutional neural network for sar image,\u201d","author":"Wang","year":"2018","journal-title":"2018 11th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI)"},{"key":"B40","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2024.3360976","article-title":"Cog-net: A cognitive network for fine-grained ship classification and retrieval in remote sensing images","volume":"62","author":"Xiong","year":"2024","journal-title":"IEEE Trans. Geosci. Remote Sens"},{"key":"B41","doi-asserted-by":"publisher","first-page":"8674","DOI":"10.1109\/TCE.2025.3586944","article-title":"SAR-GTR: Attributed scattering information guided sar graph transformer recognition algorithm for consumer-grade low-altitude economy","volume":"71","author":"Xiong","year":"2025","journal-title":"IEEE Trans. Consumer Elect"},{"key":"B42","doi-asserted-by":"publisher","first-page":"214","DOI":"10.3390\/rs17020214","article-title":"Mambashadowdet: A high-speed and high-accuracy moving target shadow detection network for video sar","volume":"17","author":"Xu","year":"2025","journal-title":"Remote Sens"},{"key":"B43","doi-asserted-by":"publisher","first-page":"2276","DOI":"10.1109\/JSTARS.2020.2991784","article-title":"Distribution shift metric learning for fine-grained ship classification in sar images","volume":"13","author":"Xu","year":"2020","journal-title":"IEEE J. Select. Topics Appl. Earth Observat. Remote Sens"},{"key":"B44","doi-asserted-by":"publisher","first-page":"2773","DOI":"10.3390\/rs17162773","article-title":"Investigating the earliest identifiable timing of sugarcane at early season based on optical and sar time-series data","volume":"17","author":"Yang","year":"2025","journal-title":"Remote Sens"},{"key":"B45","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2022.3190977","article-title":"Unsupervised sar image change detection for few changed area based on histogram fitting error minimization","volume":"60","author":"Zhang","year":"2022","journal-title":"IEEE Trans. Geosci. Remote Sens"},{"key":"B46","doi-asserted-by":"publisher","first-page":"132","DOI":"10.1109\/MGRS.2024.3404506","article-title":"Optical and synthetic aperture radar image fusion for ship detection and recognition: Current state, challenges, and future prospects","volume":"12","author":"Zhang","year":"2024","journal-title":"IEEE Geosci. Remote Sens. Magazine"},{"key":"B47","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2024.3424489","article-title":"Gradient-guided multiscale focal attention network for remote sensing scene classification","volume":"62","author":"Zhao","year":"2024","journal-title":"IEEE Trans. Geosci. Remote Sens"},{"key":"B48","doi-asserted-by":"publisher","first-page":"127070","DOI":"10.1016\/j.eswa.2025.127070","article-title":"Incorporating difference information into curriculum learning for multitemporal image classification","volume":"276","author":"Zhao","year":"2025","journal-title":"Expert Syst. Appl"},{"key":"B49","doi-asserted-by":"publisher","first-page":"576","DOI":"10.1109\/TR.2023.3309278","article-title":"A broad sparse fine-grained image classification model based on dictionary selection strategy","volume":"73","author":"Zheng","year":"2024","journal-title":"IEEE Trans. Reliabil"},{"key":"B50","doi-asserted-by":"crossref","first-page":"2921","DOI":"10.1109\/CVPR.2016.319","article-title":"\u201cLearning deep features for discriminative localization,\u201d","volume-title":"Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Zhou","year":"2016"},{"key":"B51","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2401.09417","article-title":"Vision mamba: Efficient visual representation learning with bidirectional state space model","author":"Zhu","year":"2024","journal-title":"arXiv"}],"container-title":["Frontiers in Neurorobotics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2026.1753927\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,23]],"date-time":"2026-04-23T05:36:44Z","timestamp":1776922604000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2026.1753927\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,23]]},"references-count":51,"alternative-id":["10.3389\/fnbot.2026.1753927"],"URL":"https:\/\/doi.org\/10.3389\/fnbot.2026.1753927","relation":{},"ISSN":["1662-5218"],"issn-type":[{"value":"1662-5218","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,23]]},"article-number":"1753927"}}