SpecSpatMamba: an efficient hyperspectral image classification method integrating spectral-spatial dual-path and state space model
Current hyperspectral image classification methods face three critical limitations: (1) traditional CNNs suffer from the curse of dimensionality when processing high-dimensional spectral data, leading to overfitting and poor generalization; (2) existing approaches fail to effectively address spectra...
Saved in:
| Published in: | The Egyptian journal of remote sensing and space sciences Vol. 28; no. 4; pp. 628 - 644 |
|---|---|
| Main Authors: | , |
| Format: | Journal Article |
| Language: | English |
| Published: |
Elsevier B.V
01.12.2025
|
| Subjects: | |
| ISSN: | 1110-9823 |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | Current hyperspectral image classification methods face three critical limitations: (1) traditional CNNs suffer from the curse of dimensionality when processing high-dimensional spectral data, leading to overfitting and poor generalization; (2) existing approaches fail to effectively address spectral band redundancy, resulting in computational inefficiency and suboptimal feature representation; (3) conventional methods lack synergistic utilization of spatial-spectral information, treating spectral and spatial dimensions uniformly rather than exploiting their distinct characteristics. To address these gaps, this paper proposes SpecSpatMamba, a novel hyperspectral image classification method integrating spectral-spatial dual-path feature extraction with state space models. SpecSpatMamba introduces three core innovations: (1) Dual-path feature extraction with spectral-spatial separation, where 1 × 1 convolutions extract spectral features and 3 × 3 convolutions capture spatial features; (2) Hybrid architecture combining state space models with convolutional operations for balanced long-range dependency and local feature capture; (3) Computational efficiency breakthrough achieving O(L·d) linear complexity compared to Transformer’s O(L2·d) complexity. Experiments on four benchmark datasets—Indian Pines, Pavia University, Salinas Valley, and Houston2013—demonstrate competitive performance compared to state-of-the-art methods. SpecSpatMamba achieves overall accuracies of 95.11 %, 98.61 %, 96.97 %, and 91.48 %, respectively. Notably, SpecSpatMamba demonstrates superior cross-dataset consistency and robust performance across diverse geographic environments, with particularly strong improvements in complex urban scenarios (+0.39 % on Houston2013) and agricultural settings (+0.57 % on Salinas Valley), confirming the method’s effectiveness in addressing high-dimensional hyperspectral data challenges. |
|---|---|
| AbstractList | Current hyperspectral image classification methods face three critical limitations: (1) traditional CNNs suffer from the curse of dimensionality when processing high-dimensional spectral data, leading to overfitting and poor generalization; (2) existing approaches fail to effectively address spectral band redundancy, resulting in computational inefficiency and suboptimal feature representation; (3) conventional methods lack synergistic utilization of spatial-spectral information, treating spectral and spatial dimensions uniformly rather than exploiting their distinct characteristics. To address these gaps, this paper proposes SpecSpatMamba, a novel hyperspectral image classification method integrating spectral-spatial dual-path feature extraction with state space models. SpecSpatMamba introduces three core innovations: (1) Dual-path feature extraction with spectral-spatial separation, where 1 × 1 convolutions extract spectral features and 3 × 3 convolutions capture spatial features; (2) Hybrid architecture combining state space models with convolutional operations for balanced long-range dependency and local feature capture; (3) Computational efficiency breakthrough achieving O(L·d) linear complexity compared to Transformer’s O(L2·d) complexity. Experiments on four benchmark datasets—Indian Pines, Pavia University, Salinas Valley, and Houston2013—demonstrate competitive performance compared to state-of-the-art methods. SpecSpatMamba achieves overall accuracies of 95.11 %, 98.61 %, 96.97 %, and 91.48 %, respectively. Notably, SpecSpatMamba demonstrates superior cross-dataset consistency and robust performance across diverse geographic environments, with particularly strong improvements in complex urban scenarios (+0.39 % on Houston2013) and agricultural settings (+0.57 % on Salinas Valley), confirming the method’s effectiveness in addressing high-dimensional hyperspectral data challenges. |
| Author | Liao, Jianshang Wang, Liguo |
| Author_xml | – sequence: 1 givenname: Jianshang orcidid: 0000-0003-0358-6089 surname: Liao fullname: Liao, Jianshang email: liaojianshang@126.com organization: School of Computer Science, Guangzhou Maritime University, Guangzhou 510725, China – sequence: 2 givenname: Liguo surname: Wang fullname: Wang, Liguo email: wangliguo@hrbeu.edu.cn organization: College of Information and Communications Engineering, Dalian Minzu University, Dalian 116600, China |
| BookMark | eNp9kM1OAyEUhVnUxKp9AVe8wIzAtMyMcWMa_5IaF9U1YeDSMpmfBtCkW5_cO6luZQOcnO_k3nNBZsM4ACHXnOWccXnT5tCGmAsmVijkjPEZmXPOWVZXojgnixhbhkeyVbmUc_K9PYDZHnR61X2jb6keKDjnjYch0f3xACGiIQXdUd_rHVDT6Rg9OnTy40B7SPvRUj8k2AWUhh39A7KIsR5B-4kffO8x3dKYdAI0aQO0Hy10V-TM6S7C4ve-JB-PD-_r52zz9vSyvt9kRtRlyhpZ8soJa4VzRgspayN0U9RNWVvHlxIqJiwTAIXQkpvK2mVZ10VRla6U2pbFJRGnXBPGGAM4dQi4UzgqztRUnmrVVJ6ayps0LA-huxMEONmXh6Di1I0B6wOuqezo_8N_ADsHf6U |
| Cites_doi | 10.1016/j.dt.2022.02.007 10.1117/1.JRS.16.044525 10.3390/rs11020159 10.1155/2021/9962057 10.1109/TGRS.2022.3216319 10.1109/TGRS.2023.3286950 10.1109/JSTARS.2022.3185125 10.1080/10106049.2023.2226112 10.1109/LGRS.2024.3356422 10.1109/TMM.2025.3604954 10.3390/rs16244661 10.3390/rs17152577 10.1016/j.neunet.2025.107350 10.3390/rs16071180 10.1016/j.neucom.2024.128751 10.1109/TGRS.2019.2951160 10.1109/JSTARS.2024.3461851 10.1117/1.JRS.11.046010 10.1109/TGRS.2024.3469384 10.1109/JSTARS.2021.3056124 10.3390/rs16214001 10.1109/TGRS.2024.3516817 10.1109/LGRS.2023.3255867 10.1109/TIM.2023.3279922 10.1109/TGRS.2021.3102034 10.1109/LGRS.2017.2764915 10.3390/rs15020428 10.1109/LGRS.2013.2254108 10.3390/rs13030498 10.1049/iet-ipr.2018.5063 10.1109/TGRS.2024.3493101 10.1109/LGRS.2023.3287277 10.3390/rs13132445 10.1109/LGRS.2014.2320258 10.1109/LGRS.2019.2918719 10.1109/JSTARS.2024.3510468 10.1016/j.infrared.2020.103326 10.1109/TGRS.2022.3201056 10.1155/2018/8602103 10.3390/rs12030582 10.3390/s24144760 |
| ContentType | Journal Article |
| Copyright | 2025 The Author(s) |
| Copyright_xml | – notice: 2025 The Author(s) |
| DBID | 6I. AAFTH AAYXX CITATION |
| DOI | 10.1016/j.ejrs.2025.10.001 |
| DatabaseName | ScienceDirect Open Access Titles Elsevier:ScienceDirect:Open Access CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| EndPage | 644 |
| ExternalDocumentID | 10_1016_j_ejrs_2025_10_001 S1110982325000626 |
| GroupedDBID | --K 0R~ 4.4 457 5VS 6I. AAEDT AAEDW AAFTH AAFWJ AAHBH AAIKJ AALRI AAXUO AAYWO ABMAC ACGFS ACVFH ADBBV ADCNI ADEZE ADVLN AEUPX AEXQZ AFJKZ AFPKN AFPUW AFTJW AGHFR AIGII AITUG AKBMS AKRWK AKYEP ALMA_UNASSIGNED_HOLDINGS AMRAJ APXCP BCNDV E3Z EBS EJD FDB GROUPED_DOAJ HZ~ IPNFZ IXB KQ8 M41 O-L O9- OK1 RIG ROL SES SSZ XH2 AAYXX CITATION |
| ID | FETCH-LOGICAL-c297t-b6718f2dd2ffca2669c2ab39b79df146e802d02ee32a61c8dd47993387f76ad73 |
| ISICitedReferencesCount | 0 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001598743400001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1110-9823 |
| IngestDate | Thu Nov 20 00:48:22 EST 2025 Sat Nov 08 17:26:16 EST 2025 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Issue | 4 |
| Keywords | Deep learning State space model Linear computational complexity Spectral-spatial dual-path Hyperspectral image classification |
| Language | English |
| License | This is an open access article under the CC BY license. |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c297t-b6718f2dd2ffca2669c2ab39b79df146e802d02ee32a61c8dd47993387f76ad73 |
| ORCID | 0000-0003-0358-6089 |
| OpenAccessLink | http://dx.doi.org/10.1016/j.ejrs.2025.10.001 |
| PageCount | 17 |
| ParticipantIDs | crossref_primary_10_1016_j_ejrs_2025_10_001 elsevier_sciencedirect_doi_10_1016_j_ejrs_2025_10_001 |
| PublicationCentury | 2000 |
| PublicationDate | December 2025 2025-12-00 |
| PublicationDateYYYYMMDD | 2025-12-01 |
| PublicationDate_xml | – month: 12 year: 2025 text: December 2025 |
| PublicationDecade | 2020 |
| PublicationTitle | The Egyptian journal of remote sensing and space sciences |
| PublicationYear | 2025 |
| Publisher | Elsevier B.V |
| Publisher_xml | – name: Elsevier B.V |
| References | Sun, Zheng, Lu, Wu (b0180) 2020; 58 Xing, Wang, Liu, Cheng, Xu (b0215) 2024; 16 Zhang, Bi, Hao, Du, Zhu, Gao (b0245) 2022; 16 Fang, Li, Zhang, Chan (b0065) 2019; 11 Xia, Du, He, Chanussot (b0210) 2014; 11 Mei, Li, Liu, Cai, Du (b0155) 2022; 60 Wu, Wang (b0205) 2023; 38 Liu, Wang, Liu, Li, Yang, Chen, Liu, Han (b0135) 2023; 15 Li, Luo, Zhang, Wang, Du (b0120) 2024; 62 Yang, Zhang (b0235) 2024; 24 Zhou, Zhang, Zhang, Ma (b0250) 2023; 20 He, Chen, Lin (b0095) 2021; 13 Wang, Sun, Zhang, Ren, Wang, Ren, Liu (b0195) 2025; 187 Ding, Zhang, Kang, Yang, Zhao, Feng, Xiao, Cai, Hong, Zheng (b0050) 2025; 63 Mohan, Venkatesan (b0160) 2020; 108 Ahmad, Butt, Mazzara, Distefano, Khan, Altuwaijri (b0005) 2024; 17 Yan, Zhang, Wang, Leng, Basu, Peng (b0225) 2023; 20 Ding, Y., Zhang, Z., Yang, A., Cai, Y., Xiao, X., Hong, D., & Yuan, J. (2025b). SLCGC: A lightweight self-supervised low-pass contrastive graph clustering network for hyperspectral images. arXiv preprint arXiv:2502.03497. doi:10.48550/arXiv.2502.03497. Liu, Hu, Kang, Luo, Fan (b0145) 2022; 60 Yang, Li, Ding, Fang, Cai, He (b0230) 2024; 62 Gao, Jin, Zhou, Dong, Du (b0075) 2025; 63 He, Tu, Jiang, Liu, Li, Plaza (b0100) 2024; 62 Gao, Miao, Cao, Li (b0085) 2021; 14 Chen, Zhu, Ghamisi, Jia, Li, Tang (b0025) 2017; 14 Jamali, Roy, Hong, Atkinson, Ghamisi (b0105) 2024; 21 Ding, Zhang, Zhao, Cai, Yang, Hu, Xu, Li, Cai, Cai (b0060) 2022; 60 Cheng, Chan, Du (b0030) 2024; 62 Liu, R. M., Ning, X., Cai, W. W., & Li, G. J. (2021). Multiscale Dense Cross-Attention Mechanism with Covariance Pooling for Hyperspectral Image Scene Classification. MOBILE INFORMATION SYSTEMS, 2021, 9962057. doi:10.1155/2021/9962057. Ren, Tu, Li, He, Peng (b0165) 2022; 15 Xu, Xue, Li, Cheng, Su, Xia (b0220) 2025; 63 Gao, H. M., Lin, S., Yang, Y., Li, C. M., & Yang, M. X. (2018). Convolution Neural Network Based on Two-Dimensional Spectrum for Hyperspectral Image Classification. JOURNAL OF SENSORS, 2018, 8602103. doi:10.1155/2018/8602103. Sheng, Zhou, Wang, Ye, Fan (b0175) 2025; 63 Aria, Menenti, Gorte (b0015) 2017; 11 Anandakrishnan, Sundaram, Paneer (b0010) 2025; 18 Roy, Krishna, Dubey, Chaudhuri (b0170) 2020; 17 Wang, Yang, Cui, Ding, Xue, Su (b0190) 2024; 16 Yao, Zhang, Zhao, Cai, He, Cai, Cai (b0240) 2023; 23 Wang, Sun, Lu, Li (b0200) 2024; 16 Ding, Li, Yang, Li, Liu, Liu, Zhang (b0045) 2021; 13 Fu, Xiong, Lu, Zhou, Zhou, Qian (b0070) 2024; 62 Liao, Shi, Wang (b0125) 2023; 61 Cui, Xia, Wang, Gao, Wang (b0035) 2022; 60 Hamouda, Ettabaa, Bouhlel (b0090) 2019; 13 Cai, Ning, Zhou, Bai, Jiang, Li, Qian (b0020) 2023; 61 Khodadadzadeh, Li, Plaza, Bioucas-Dias (b0110) 2014; 11 Liao, Wang (b0130) 2025; 17 Wang, Wang, Wu (b0185) 2025; 20 Ding, Lu, Fu, Li, Ma (b0040) 2022; 60 Zhou, Kamata, Wang, Wong, Hou (b0255) 2025; 613 Li, Zheng, Duan, Yang, Wang (b0115) 2020; 12 Ma, Wan, Wu, Kong, Shao, Wang, Chen, Gu (b0150) 2023; 72 Ahmad (10.1016/j.ejrs.2025.10.001_b0005) 2024; 17 He (10.1016/j.ejrs.2025.10.001_b0095) 2021; 13 Chen (10.1016/j.ejrs.2025.10.001_b0025) 2017; 14 Li (10.1016/j.ejrs.2025.10.001_b0120) 2024; 62 Fu (10.1016/j.ejrs.2025.10.001_b0070) 2024; 62 Ding (10.1016/j.ejrs.2025.10.001_b0050) 2025; 63 Ding (10.1016/j.ejrs.2025.10.001_b0040) 2022; 60 Sheng (10.1016/j.ejrs.2025.10.001_b0175) 2025; 63 Yao (10.1016/j.ejrs.2025.10.001_b0240) 2023; 23 Liu (10.1016/j.ejrs.2025.10.001_b0135) 2023; 15 10.1016/j.ejrs.2025.10.001_b0140 Ren (10.1016/j.ejrs.2025.10.001_b0165) 2022; 15 Jamali (10.1016/j.ejrs.2025.10.001_b0105) 2024; 21 Xia (10.1016/j.ejrs.2025.10.001_b0210) 2014; 11 Sun (10.1016/j.ejrs.2025.10.001_b0180) 2020; 58 Gao (10.1016/j.ejrs.2025.10.001_b0085) 2021; 14 Hamouda (10.1016/j.ejrs.2025.10.001_b0090) 2019; 13 Xing (10.1016/j.ejrs.2025.10.001_b0215) 2024; 16 10.1016/j.ejrs.2025.10.001_b0055 He (10.1016/j.ejrs.2025.10.001_b0100) 2024; 62 Liao (10.1016/j.ejrs.2025.10.001_b0125) 2023; 61 Zhou (10.1016/j.ejrs.2025.10.001_b0250) 2023; 20 Liu (10.1016/j.ejrs.2025.10.001_b0145) 2022; 60 Wang (10.1016/j.ejrs.2025.10.001_b0185) 2025; 20 Khodadadzadeh (10.1016/j.ejrs.2025.10.001_b0110) 2014; 11 Yang (10.1016/j.ejrs.2025.10.001_b0235) 2024; 24 Zhang (10.1016/j.ejrs.2025.10.001_b0245) 2022; 16 Zhou (10.1016/j.ejrs.2025.10.001_b0255) 2025; 613 Ding (10.1016/j.ejrs.2025.10.001_b0060) 2022; 60 Gao (10.1016/j.ejrs.2025.10.001_b0075) 2025; 63 Ding (10.1016/j.ejrs.2025.10.001_b0045) 2021; 13 Cheng (10.1016/j.ejrs.2025.10.001_b0030) 2024; 62 Fang (10.1016/j.ejrs.2025.10.001_b0065) 2019; 11 10.1016/j.ejrs.2025.10.001_b0080 Wang (10.1016/j.ejrs.2025.10.001_b0195) 2025; 187 Li (10.1016/j.ejrs.2025.10.001_b0115) 2020; 12 Wang (10.1016/j.ejrs.2025.10.001_b0200) 2024; 16 Roy (10.1016/j.ejrs.2025.10.001_b0170) 2020; 17 Ma (10.1016/j.ejrs.2025.10.001_b0150) 2023; 72 Mei (10.1016/j.ejrs.2025.10.001_b0155) 2022; 60 Aria (10.1016/j.ejrs.2025.10.001_b0015) 2017; 11 Anandakrishnan (10.1016/j.ejrs.2025.10.001_b0010) 2025; 18 Cai (10.1016/j.ejrs.2025.10.001_b0020) 2023; 61 Liao (10.1016/j.ejrs.2025.10.001_b0130) 2025; 17 Wang (10.1016/j.ejrs.2025.10.001_b0190) 2024; 16 Wu (10.1016/j.ejrs.2025.10.001_b0205) 2023; 38 Yan (10.1016/j.ejrs.2025.10.001_b0225) 2023; 20 Xu (10.1016/j.ejrs.2025.10.001_b0220) 2025; 63 Cui (10.1016/j.ejrs.2025.10.001_b0035) 2022; 60 Mohan (10.1016/j.ejrs.2025.10.001_b0160) 2020; 108 Yang (10.1016/j.ejrs.2025.10.001_b0230) 2024; 62 |
| References_xml | – volume: 11 start-page: 2105 year: 2014 end-page: 2109 ident: b0110 article-title: A subspace-based multinomial logistic regression for hyperspectral image classification publication-title: IEEE Geosci. Remote Sens. Lett. – volume: 11 start-page: 239 year: 2014 end-page: 243 ident: b0210 article-title: Hyperspectral remote sensing image classification based on rotation forest publication-title: IEEE Geosci. Remote Sens. Lett. – volume: 16 year: 2022 ident: b0245 article-title: Transformer attention network and unmanned aerial vehicle hyperspectral remote sensing for grassland rodent pest monitoring research publication-title: J. Appl. Remote Sens. – volume: 11 start-page: 159 year: 2019 ident: b0065 article-title: Hyperspectral images classification based on dense convolutional networks with spectral-wise attention mechanism publication-title: Remote Sens. (Basel) – volume: 60 year: 2022 ident: b0040 article-title: Global-local transformer network for HSI and LiDAR data joint classification publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 63 year: 2025 ident: b0175 article-title: DualMamba: a lightweight spectral-spatial mamba-convolution network for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 62 year: 2024 ident: b0120 article-title: MambaHSI: spatial-spectral mamba for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 17 start-page: 277 year: 2020 end-page: 281 ident: b0170 article-title: HybridSN: exploring 3-D-2-D CNN feature hierarchy for hyperspectral image classification publication-title: IEEE Geosci. Remote Sens. Lett. – volume: 61 year: 2023 ident: b0125 article-title: A spectral-spatial fusion transformer network for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 16 start-page: 4001 year: 2024 ident: b0190 article-title: Capsule attention network for hyperspectral image classification publication-title: Remote Sens. (Basel) – volume: 14 start-page: 2355 year: 2017 end-page: 2359 ident: b0025 article-title: Hyperspectral images classification with gabor filtering and convolutional neural network publication-title: IEEE Geosci. Remote Sens. Lett. – volume: 15 start-page: 428 year: 2023 ident: b0135 article-title: A Multiscale cross interaction attention network for hyperspectral image classification publication-title: Remote Sens. (Basel) – volume: 11 year: 2017 ident: b0015 article-title: Spectral region identification versus individual channel selection in supervised dimensionality reduction of hyperspectral image data publication-title: J. Appl. Remote Sens. – volume: 613 year: 2025 ident: b0255 article-title: Mamba-in-mamba: centralized mamba-cross-scan in tokenized mamba model for hyperspectral image classification publication-title: Neurocomputing – volume: 187 year: 2025 ident: b0195 article-title: A spatial-spectral fusion convolutional transformer network with contextual multi-head self-attention for hyperspectral image classification publication-title: Neural Netw. – volume: 21 year: 2024 ident: b0105 article-title: Attention graph convolutional network for disjoint hyperspectral image classification publication-title: IEEE Geosci. Remote Sens. Lett. – reference: Liu, R. M., Ning, X., Cai, W. W., & Li, G. J. (2021). Multiscale Dense Cross-Attention Mechanism with Covariance Pooling for Hyperspectral Image Scene Classification. MOBILE INFORMATION SYSTEMS, 2021, 9962057. doi:10.1155/2021/9962057. – volume: 62 year: 2024 ident: b0070 article-title: Hyperspectral image denoising via spatial-spectral recurrent transformer publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 62 year: 2024 ident: b0030 article-title: MS2I2Former: multiscale spatial–spectral information interactive transformer for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 23 start-page: 164 year: 2023 end-page: 176 ident: b0240 article-title: Deep hybrid: Multi-graph neural network collaboration for hyperspectral image classification publication-title: Def. Technol. – volume: 60 year: 2022 ident: b0145 article-title: Interactformer: interactive transformer and CNN for hyperspectral image super-resolution publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 58 start-page: 3232 year: 2020 end-page: 3245 ident: b0180 article-title: Spectral-spatial attention network for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 72 year: 2023 ident: b0150 article-title: Light self-gaussian-attention vision transformer for hyperspectral image classification publication-title: IEEE Trans. Instrum. Meas. – volume: 63 year: 2025 ident: b0220 article-title: UM2Former: U-shaped multimixed transformer network for large-scale hyperspectral image semantic segmentation publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 108 year: 2020 ident: b0160 article-title: HybridCNN based hyperspectral image classification using multiscale spatiospectral features publication-title: Infrared Phys. Technol. – volume: 13 start-page: 392 year: 2019 end-page: 398 ident: b0090 article-title: Hyperspectral imaging classification based on convolutional neural networks by adaptive sizes of windows and filters publication-title: IET Image Proc. – volume: 62 year: 2024 ident: b0100 article-title: IGroupSS-Mamba: interval group spatial-spectral mamba for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 17 start-page: 2577 year: 2025 ident: b0130 article-title: HyperspectralMamba: a novel state space model architecture for hyperspectral image classification publication-title: Remote Sens. (Basel) – volume: 61 year: 2023 ident: b0020 article-title: A novel hyperspectral image classification model using bole convolution with three-direction attention mechanism: small sample and unbalanced learning publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 20 year: 2023 ident: b0225 article-title: Hybrid conv-ViT Network for hyperspectral image classification publication-title: IEEE Geosci. Remote Sens. Lett. – volume: 63 year: 2025 ident: b0075 article-title: MSFMamba: multiscale feature fusion state space model for multisource remote sensing image classification publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 20 year: 2023 ident: b0250 article-title: Vision Transformer with contrastive learning for hyperspectral image classification publication-title: IEEE Geosci. Remote Sens. Lett. – volume: 12 start-page: 582 year: 2020 ident: b0115 article-title: Classification of hyperspectral image based on double-branch dual-attention mechanism network publication-title: Remote Sens. (Basel) – volume: 13 start-page: 498 year: 2021 ident: b0095 article-title: Spatial-spectral transformer for hyperspectral image classification publication-title: Remote Sens. (Basel) – volume: 18 start-page: 1817 year: 2025 end-page: 1826 ident: b0010 article-title: STA-AgriNet: a spatio-temporal attention framework for crop type mapping from fused multi-sensor multi-temporal SITS publication-title: IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. – volume: 15 start-page: 5115 year: 2022 end-page: 5130 ident: b0165 article-title: Multiscale adaptive convolution for hyperspectral image classification publication-title: IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. – volume: 60 year: 2022 ident: b0155 article-title: Hyperspectral image classification using attention-based bidirectional long short-term memory network publication-title: IEEE Trans. Geosci. Remote Sens. – reference: Gao, H. M., Lin, S., Yang, Y., Li, C. M., & Yang, M. X. (2018). Convolution Neural Network Based on Two-Dimensional Spectrum for Hyperspectral Image Classification. JOURNAL OF SENSORS, 2018, 8602103. doi:10.1155/2018/8602103. – volume: 16 start-page: 1180 year: 2024 ident: b0200 article-title: A novel Transformer network with a CNN-enhanced cross-attention mechanism for hyperspectral image classification publication-title: Remote Sens. (Basel) – volume: 20 year: 2025 ident: b0185 article-title: TGF-Net: transformer and gist CNN fusion network for multi-modal remote sensing image classification publication-title: PLoS One – volume: 16 start-page: 4661 year: 2024 ident: b0215 article-title: MambaHR: state space model for hyperspectral image restoration under stray light interference publication-title: Remote Sens. (Basel) – volume: 62 year: 2024 ident: b0230 article-title: GraphMamba: an efficient graph structure learning vision mamba for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 60 year: 2022 ident: b0035 article-title: Lightweight spectral-spatial attention network for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 60 year: 2022 ident: b0060 article-title: Unsupervised self-correlated learning smoothy enhanced locality preserving graph convolution embedding clustering for hyperspectral images publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 63 start-page: 1 year: 2025 end-page: 13 ident: b0050 article-title: Adaptive homophily clustering: structure homophily graph learning with adaptive filter for hyperspectral image publication-title: IEEE Trans. Geosci. Remote Sens. – reference: Ding, Y., Zhang, Z., Yang, A., Cai, Y., Xiao, X., Hong, D., & Yuan, J. (2025b). SLCGC: A lightweight self-supervised low-pass contrastive graph clustering network for hyperspectral images. arXiv preprint arXiv:2502.03497. doi:10.48550/arXiv.2502.03497. – volume: 17 start-page: 17681 year: 2024 end-page: 17689 ident: b0005 article-title: Pyramid hierarchical spatial-spectral transformer for hyperspectral image classification publication-title: IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. – volume: 24 start-page: 4760 year: 2024 ident: b0235 article-title: A dual-branch fusion of a graph convolutional network and a convolutional neural network for hyperspectral image classification publication-title: Sensors – volume: 14 start-page: 2563 year: 2021 end-page: 2576 ident: b0085 article-title: Densely connected multiscale attention network for hyperspectral image classification publication-title: IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. – volume: 13 start-page: 2445 year: 2021 ident: b0045 article-title: An adaptive capsule network for hyperspectral remote sensing classification publication-title: Remote Sens. (Basel) – volume: 38 year: 2023 ident: b0205 article-title: Global and pyramid convolutional neural network with hybrid attention mechanism for hyperspectral image classification publication-title: Geocarto Int. – volume: 63 start-page: 1 year: 2025 ident: 10.1016/j.ejrs.2025.10.001_b0050 article-title: Adaptive homophily clustering: structure homophily graph learning with adaptive filter for hyperspectral image publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 62 year: 2024 ident: 10.1016/j.ejrs.2025.10.001_b0120 article-title: MambaHSI: spatial-spectral mamba for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 23 start-page: 164 year: 2023 ident: 10.1016/j.ejrs.2025.10.001_b0240 article-title: Deep hybrid: Multi-graph neural network collaboration for hyperspectral image classification publication-title: Def. Technol. doi: 10.1016/j.dt.2022.02.007 – volume: 16 issue: 4 year: 2022 ident: 10.1016/j.ejrs.2025.10.001_b0245 article-title: Transformer attention network and unmanned aerial vehicle hyperspectral remote sensing for grassland rodent pest monitoring research publication-title: J. Appl. Remote Sens. doi: 10.1117/1.JRS.16.044525 – volume: 11 start-page: 159 issue: 2 year: 2019 ident: 10.1016/j.ejrs.2025.10.001_b0065 article-title: Hyperspectral images classification based on dense convolutional networks with spectral-wise attention mechanism publication-title: Remote Sens. (Basel) doi: 10.3390/rs11020159 – ident: 10.1016/j.ejrs.2025.10.001_b0140 doi: 10.1155/2021/9962057 – volume: 60 year: 2022 ident: 10.1016/j.ejrs.2025.10.001_b0040 article-title: Global-local transformer network for HSI and LiDAR data joint classification publication-title: IEEE Trans. Geosci. Remote Sens. doi: 10.1109/TGRS.2022.3216319 – volume: 60 year: 2022 ident: 10.1016/j.ejrs.2025.10.001_b0145 article-title: Interactformer: interactive transformer and CNN for hyperspectral image super-resolution publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 61 year: 2023 ident: 10.1016/j.ejrs.2025.10.001_b0125 article-title: A spectral-spatial fusion transformer network for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. doi: 10.1109/TGRS.2023.3286950 – volume: 15 start-page: 5115 year: 2022 ident: 10.1016/j.ejrs.2025.10.001_b0165 article-title: Multiscale adaptive convolution for hyperspectral image classification publication-title: IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. doi: 10.1109/JSTARS.2022.3185125 – volume: 38 issue: 1 year: 2023 ident: 10.1016/j.ejrs.2025.10.001_b0205 article-title: Global and pyramid convolutional neural network with hybrid attention mechanism for hyperspectral image classification publication-title: Geocarto Int. doi: 10.1080/10106049.2023.2226112 – volume: 20 issue: 2 year: 2025 ident: 10.1016/j.ejrs.2025.10.001_b0185 article-title: TGF-Net: transformer and gist CNN fusion network for multi-modal remote sensing image classification publication-title: PLoS One – volume: 21 year: 2024 ident: 10.1016/j.ejrs.2025.10.001_b0105 article-title: Attention graph convolutional network for disjoint hyperspectral image classification publication-title: IEEE Geosci. Remote Sens. Lett. doi: 10.1109/LGRS.2024.3356422 – ident: 10.1016/j.ejrs.2025.10.001_b0055 doi: 10.1109/TMM.2025.3604954 – volume: 16 start-page: 4661 issue: 24 year: 2024 ident: 10.1016/j.ejrs.2025.10.001_b0215 article-title: MambaHR: state space model for hyperspectral image restoration under stray light interference publication-title: Remote Sens. (Basel) doi: 10.3390/rs16244661 – volume: 63 year: 2025 ident: 10.1016/j.ejrs.2025.10.001_b0220 article-title: UM2Former: U-shaped multimixed transformer network for large-scale hyperspectral image semantic segmentation publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 17 start-page: 2577 issue: 15 year: 2025 ident: 10.1016/j.ejrs.2025.10.001_b0130 article-title: HyperspectralMamba: a novel state space model architecture for hyperspectral image classification publication-title: Remote Sens. (Basel) doi: 10.3390/rs17152577 – volume: 187 year: 2025 ident: 10.1016/j.ejrs.2025.10.001_b0195 article-title: A spatial-spectral fusion convolutional transformer network with contextual multi-head self-attention for hyperspectral image classification publication-title: Neural Netw. doi: 10.1016/j.neunet.2025.107350 – volume: 16 start-page: 1180 issue: 7 year: 2024 ident: 10.1016/j.ejrs.2025.10.001_b0200 article-title: A novel Transformer network with a CNN-enhanced cross-attention mechanism for hyperspectral image classification publication-title: Remote Sens. (Basel) doi: 10.3390/rs16071180 – volume: 613 year: 2025 ident: 10.1016/j.ejrs.2025.10.001_b0255 article-title: Mamba-in-mamba: centralized mamba-cross-scan in tokenized mamba model for hyperspectral image classification publication-title: Neurocomputing doi: 10.1016/j.neucom.2024.128751 – volume: 58 start-page: 3232 issue: 5 year: 2020 ident: 10.1016/j.ejrs.2025.10.001_b0180 article-title: Spectral-spatial attention network for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. doi: 10.1109/TGRS.2019.2951160 – volume: 60 year: 2022 ident: 10.1016/j.ejrs.2025.10.001_b0060 article-title: Unsupervised self-correlated learning smoothy enhanced locality preserving graph convolution embedding clustering for hyperspectral images publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 17 start-page: 17681 year: 2024 ident: 10.1016/j.ejrs.2025.10.001_b0005 article-title: Pyramid hierarchical spatial-spectral transformer for hyperspectral image classification publication-title: IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. doi: 10.1109/JSTARS.2024.3461851 – volume: 11 year: 2017 ident: 10.1016/j.ejrs.2025.10.001_b0015 article-title: Spectral region identification versus individual channel selection in supervised dimensionality reduction of hyperspectral image data publication-title: J. Appl. Remote Sens. doi: 10.1117/1.JRS.11.046010 – volume: 62 year: 2024 ident: 10.1016/j.ejrs.2025.10.001_b0030 article-title: MS2I2Former: multiscale spatial–spectral information interactive transformer for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. doi: 10.1109/TGRS.2024.3469384 – volume: 14 start-page: 2563 year: 2021 ident: 10.1016/j.ejrs.2025.10.001_b0085 article-title: Densely connected multiscale attention network for hyperspectral image classification publication-title: IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. doi: 10.1109/JSTARS.2021.3056124 – volume: 16 start-page: 4001 issue: 21 year: 2024 ident: 10.1016/j.ejrs.2025.10.001_b0190 article-title: Capsule attention network for hyperspectral image classification publication-title: Remote Sens. (Basel) doi: 10.3390/rs16214001 – volume: 63 year: 2025 ident: 10.1016/j.ejrs.2025.10.001_b0175 article-title: DualMamba: a lightweight spectral-spatial mamba-convolution network for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. doi: 10.1109/TGRS.2024.3516817 – volume: 20 year: 2023 ident: 10.1016/j.ejrs.2025.10.001_b0250 article-title: Vision Transformer with contrastive learning for hyperspectral image classification publication-title: IEEE Geosci. Remote Sens. Lett. doi: 10.1109/LGRS.2023.3255867 – volume: 72 year: 2023 ident: 10.1016/j.ejrs.2025.10.001_b0150 article-title: Light self-gaussian-attention vision transformer for hyperspectral image classification publication-title: IEEE Trans. Instrum. Meas. doi: 10.1109/TIM.2023.3279922 – volume: 60 year: 2022 ident: 10.1016/j.ejrs.2025.10.001_b0155 article-title: Hyperspectral image classification using attention-based bidirectional long short-term memory network publication-title: IEEE Trans. Geosci. Remote Sens. doi: 10.1109/TGRS.2021.3102034 – volume: 14 start-page: 2355 issue: 12 year: 2017 ident: 10.1016/j.ejrs.2025.10.001_b0025 article-title: Hyperspectral images classification with gabor filtering and convolutional neural network publication-title: IEEE Geosci. Remote Sens. Lett. doi: 10.1109/LGRS.2017.2764915 – volume: 62 year: 2024 ident: 10.1016/j.ejrs.2025.10.001_b0070 article-title: Hyperspectral image denoising via spatial-spectral recurrent transformer publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 63 year: 2025 ident: 10.1016/j.ejrs.2025.10.001_b0075 article-title: MSFMamba: multiscale feature fusion state space model for multisource remote sensing image classification publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 60 year: 2022 ident: 10.1016/j.ejrs.2025.10.001_b0035 article-title: Lightweight spectral-spatial attention network for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 15 start-page: 428 issue: 2 year: 2023 ident: 10.1016/j.ejrs.2025.10.001_b0135 article-title: A Multiscale cross interaction attention network for hyperspectral image classification publication-title: Remote Sens. (Basel) doi: 10.3390/rs15020428 – volume: 11 start-page: 239 issue: 1 year: 2014 ident: 10.1016/j.ejrs.2025.10.001_b0210 article-title: Hyperspectral remote sensing image classification based on rotation forest publication-title: IEEE Geosci. Remote Sens. Lett. doi: 10.1109/LGRS.2013.2254108 – volume: 13 start-page: 498 issue: 3 year: 2021 ident: 10.1016/j.ejrs.2025.10.001_b0095 article-title: Spatial-spectral transformer for hyperspectral image classification publication-title: Remote Sens. (Basel) doi: 10.3390/rs13030498 – volume: 13 start-page: 392 issue: 2 year: 2019 ident: 10.1016/j.ejrs.2025.10.001_b0090 article-title: Hyperspectral imaging classification based on convolutional neural networks by adaptive sizes of windows and filters publication-title: IET Image Proc. doi: 10.1049/iet-ipr.2018.5063 – volume: 62 year: 2024 ident: 10.1016/j.ejrs.2025.10.001_b0230 article-title: GraphMamba: an efficient graph structure learning vision mamba for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. doi: 10.1109/TGRS.2024.3493101 – volume: 20 year: 2023 ident: 10.1016/j.ejrs.2025.10.001_b0225 article-title: Hybrid conv-ViT Network for hyperspectral image classification publication-title: IEEE Geosci. Remote Sens. Lett. doi: 10.1109/LGRS.2023.3287277 – volume: 13 start-page: 2445 issue: 13 year: 2021 ident: 10.1016/j.ejrs.2025.10.001_b0045 article-title: An adaptive capsule network for hyperspectral remote sensing classification publication-title: Remote Sens. (Basel) doi: 10.3390/rs13132445 – volume: 11 start-page: 2105 issue: 12 year: 2014 ident: 10.1016/j.ejrs.2025.10.001_b0110 article-title: A subspace-based multinomial logistic regression for hyperspectral image classification publication-title: IEEE Geosci. Remote Sens. Lett. doi: 10.1109/LGRS.2014.2320258 – volume: 17 start-page: 277 issue: 2 year: 2020 ident: 10.1016/j.ejrs.2025.10.001_b0170 article-title: HybridSN: exploring 3-D-2-D CNN feature hierarchy for hyperspectral image classification publication-title: IEEE Geosci. Remote Sens. Lett. doi: 10.1109/LGRS.2019.2918719 – volume: 18 start-page: 1817 year: 2025 ident: 10.1016/j.ejrs.2025.10.001_b0010 article-title: STA-AgriNet: a spatio-temporal attention framework for crop type mapping from fused multi-sensor multi-temporal SITS publication-title: IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. doi: 10.1109/JSTARS.2024.3510468 – volume: 62 year: 2024 ident: 10.1016/j.ejrs.2025.10.001_b0100 article-title: IGroupSS-Mamba: interval group spatial-spectral mamba for hyperspectral image classification publication-title: IEEE Trans. Geosci. Remote Sens. – volume: 108 year: 2020 ident: 10.1016/j.ejrs.2025.10.001_b0160 article-title: HybridCNN based hyperspectral image classification using multiscale spatiospectral features publication-title: Infrared Phys. Technol. doi: 10.1016/j.infrared.2020.103326 – volume: 61 year: 2023 ident: 10.1016/j.ejrs.2025.10.001_b0020 article-title: A novel hyperspectral image classification model using bole convolution with three-direction attention mechanism: small sample and unbalanced learning publication-title: IEEE Trans. Geosci. Remote Sens. doi: 10.1109/TGRS.2022.3201056 – ident: 10.1016/j.ejrs.2025.10.001_b0080 doi: 10.1155/2018/8602103 – volume: 12 start-page: 582 issue: 3 year: 2020 ident: 10.1016/j.ejrs.2025.10.001_b0115 article-title: Classification of hyperspectral image based on double-branch dual-attention mechanism network publication-title: Remote Sens. (Basel) doi: 10.3390/rs12030582 – volume: 24 start-page: 4760 issue: 14 year: 2024 ident: 10.1016/j.ejrs.2025.10.001_b0235 article-title: A dual-branch fusion of a graph convolutional network and a convolutional neural network for hyperspectral image classification publication-title: Sensors doi: 10.3390/s24144760 |
| SSID | ssj0000605746 |
| Score | 2.3354142 |
| Snippet | Current hyperspectral image classification methods face three critical limitations: (1) traditional CNNs suffer from the curse of dimensionality when... |
| SourceID | crossref elsevier |
| SourceType | Index Database Publisher |
| StartPage | 628 |
| SubjectTerms | Deep learning Hyperspectral image classification Linear computational complexity Spectral-spatial dual-path State space model |
| Title | SpecSpatMamba: an efficient hyperspectral image classification method integrating spectral-spatial dual-path and state space model |
| URI | https://dx.doi.org/10.1016/j.ejrs.2025.10.001 |
| Volume | 28 |
| WOSCitedRecordID | wos001598743400001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVAON databaseName: DOAJ Directory of Open Access Journals issn: 1110-9823 databaseCode: DOA dateStart: 20100101 customDbUrl: isFulltext: true dateEnd: 99991231 titleUrlDefault: https://www.doaj.org/ omitProxy: false ssIdentifier: ssj0000605746 providerName: Directory of Open Access Journals |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1LTxsxELZS4NALatVW5VHkQ2-VEfE-bPeGKioOFFUqSNxW3rUXEpUF5YE480P4rZ3xeDebUKG2Ui-rxIntyPPF83k8D8Y-YmxkktW5QG0kQF87UdrMCtCuDjWaK10dik2o01N9cWG-DwaPbSzM3U_VNPr-3tz-V1FDGwgbQ2f_QtzdoNAAr0Ho8ASxw_OPBI8V5bHQ8Dd7XYbCO_AP9iFRBF77X8G5k8IrQ7qNa3TZqZBBo8sQgYGKSneJJILBIXYQU3TAxnudObzBasbh8iFEJcGXLGwSobROn_IiEI-w2Nko2vwj_514QAn0Qg_6GChJI0Sl3JH9k5Gl6yEYYIr27cUlAG1UJ6PL-U3ffCGzFVeQp3E1YRsGUiKMpkjkdp-WuofHtLfp5vET0t855ZN8ohrISjHe9-MJpmmX2X7w6hsuFGHnnvgD58fpJZaLgCPfC7YuVWZ078hOmh7YLgWwtT84xmWRC-HqVL_nPj0-c_aKbcaDCD8kAL1mA9-8YQ9L4PnMbcM76PAl6PAAHb4MHU7Q4T3o8FXo8A46MLrjATo8CJ4H6Lxl51-Pzr4ci1ilQ1TSqJkoc6A3tXRO1nVlge-ZStoyMaUyrgY97PWBdAfS-0TafFhp51IFpDjRqla5dSp5x9aam8a_Z7xKa115IKjQPbVqaIA-5d4MvVbep0m6xT6161fcUjKWovVSHBe42gWuNrbBam-xrF3iIiKXaGIBkHim3_Y_9tthLxcA32Vrs8ncf2Ab1d1sNJ3sBeT8AneaoGQ |
| linkProvider | Directory of Open Access Journals |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=SpecSpatMamba%3A+an+efficient+hyperspectral+image+classification+method+integrating+spectral-spatial+dual-path+and+state+space+model&rft.jtitle=The+Egyptian+journal+of+remote+sensing+and+space+sciences&rft.au=Liao%2C+Jianshang&rft.au=Wang%2C+Liguo&rft.date=2025-12-01&rft.pub=Elsevier+B.V&rft.issn=1110-9823&rft.volume=28&rft.issue=4&rft.spage=628&rft.epage=644&rft_id=info:doi/10.1016%2Fj.ejrs.2025.10.001&rft.externalDocID=S1110982325000626 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1110-9823&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1110-9823&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1110-9823&client=summon |