基于SE注意力增强的双分支子像素引导高光谱图像分类网络
SE-Aware Dual-Branch Subpixel-Guided Network for Hyperspectral Image Classification
摘要: 高光谱图像(Hyperspectral Image, HSI)富含光谱与空间信息，在农业、环境等领域应用广泛，但高维冗余、类别不均衡和混合像元等问题使传统分类易出现混淆与误判。本文在DSNet基础上提出SE-DSNet，通过空间–光谱特征分支与子像素解混分支联合建模空间与丰度信息，增强混合像元表达；引入SE通道注意力自适应加权关键波段，结合Focal Loss缓解类别不平衡。在Indian Pines与Augsburg数据集上与CNN2D、CNN3D、GRU及DSNet对比，实验表明SE-DSNet在OA、AA和Kappa上均取得较优结果，验证了其在复杂场景下的有效性与泛化能力。
Abstract: Hyperspectral images (HSIs) are rich in spectral and spatial information and are widely used in agriculture, environmental monitoring, and other fields. However, problems such as high-dimensional redundancy, class imbalance, and mixed pixels tend to cause confusion and mis-classification in traditional classification. This paper proposes SE-DSNet based on DSNet, which jointly models spatial and abundance information through a spatial-spectral feature branch and a subpixel unmixing branch to enhance the representation of mixed pixels. It introduces SE channel attention to adaptively weight key bands and combines Focal Loss to alleviate class imbalance. Experiments conducted on the Indian Pines and Augsburg datasets demonstrate that SE-DSNet outperforms CNN2D, CNN3D, GRU, and DSNet in terms of overall accuracy (OA), average accuracy (AA), and Kappa coefficient, verifying its effectiveness and generalization capability in complex scenarios.
文章引用：郑博文. 基于SE注意力增强的双分支子像素引导高光谱图像分类网络[J]. 图像与信号处理, 2026, 15(4): 605-621. https://doi.org/10.12677/jisp.2026.154053

参考文献

[1] Chen, Y., Lin, Z., Zhao, X., Wang, G. and Gu, Y. (2014) Deep Learning-Based Classification of Hyperspectral Data: A Comprehensive Review. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 7, 2094-2107.
https://doi.org/10.1109/jstars.2014.2329330
[2] Wu, H. and Prasad, S. (2018) Semi-Supervised Deep Learning Using Pseudo Labels for Hyperspectral Image Classification. IEEE Transactions on Image Processing, 27, 1259-1270.
https://doi.org/10.1109/tip.2017.2772836
[3] Zhang, L., Zhang, L. and Du, B. (2016) Deep Learning for Remote Sensing Data: A Technical Tutorial on the State of the Art. IEEE Geoscience and Remote Sensing Magazine, 4, 22-40.
https://doi.org/10.1109/mgrs.2016.2540798
[4] Roy, S.K., Krishna, G., Dubey, S.R. and Chaudhuri, B.B. (2020) Hybridsn: Exploring 3-D-2-D CNN Feature Hierarchy for Hyperspectral Image Classification. IEEE Geoscience and Remote Sensing Letters, 17, 277-281.
https://doi.org/10.1109/lgrs.2019.2918719
[5] Hang, R., Liu, Q., Hong, D. and Ghamisi, P. (2019) Cascaded Recurrent Neural Networks for Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing, 57, 5384-5394.
https://doi.org/10.1109/tgrs.2019.2899129
[6] Ghaderizadeh, S., Abbasi-Moghadam, D., Sharifi, A., Tariq, A. and Qin, S. (2022) Multiscale Dual-Branch Residual Spectral-Spatial Network with Attention for Hyperspectral Image Classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 15, 5455-5467.
https://doi.org/10.1109/jstars.2022.3188732
[7] Rasti, B., Hong, D., Hang, R., Ghamisi, P., Kang, X., Chanussot, J., et al. (2020) Feature Extraction for Hyperspectral Imagery: The Evolution from Shallow to Deep: Overview and Toolbox. IEEE Geoscience and Remote Sensing Magazine, 8, 60-88.
https://doi.org/10.1109/mgrs.2020.2979764
[8] Li, W., Wu, G., Zhang, F. and Du, Q. (2017) Hyperspectral Image Classification Using Deep Pixel-Pair Features. IEEE Transactions on Geoscience and Remote Sensing, 55, 844-853.
https://doi.org/10.1109/tgrs.2016.2616355
[9] Yang, X., Ye, Y., Li, X., Lau, R.Y.K., Zhang, X. and Huang, X. (2018) Hyperspectral Image Classification with Deep Learning Models. IEEE Transactions on Geoscience and Remote Sensing, 56, 5408-5423.
https://doi.org/10.1109/tgrs.2018.2815613
[10] He, K., Zhang, X., Ren, S. and Sun, J. (2016) Deep Residual Learning for Image Recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, 27-30 June 2016, 770-778.
https://doi.org/10.1109/cvpr.2016.90
[11] Zhang, L., Wang, Y., Yang, L.Z., et al. (2023) A Multi-Range Spectral-Spatial Transformer for Hyperspectral Image Classification. Infrared Physics & Technology, 135, Article ID: 104983.
https://doi.org/10.1016/j.infrared.2023.104983
[12] Hong, D., Gao, L., Yao, J., Zhang, B., Plaza, A. and Chanussot, J. (2021) Graph Convolutional Networks for Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing, 59, 5966-5978.
https://doi.org/10.1109/tgrs.2020.3015157
[13] Jia, S., Jiang, S., Lin, Z., Li, N., Xu, M. and Yu, S. (2021) A Survey: Deep Learning for Hyperspectral Image Classification with Few Labeled Samples. Neurocomputing, 448, 179-204.
https://doi.org/10.1016/j.neucom.2021.03.035
[14] Dobigeon, N., Tourneret, J., Richard, C., Bermudez, J.C.M., McLaughlin, S. and Hero, A.O. (2014) Nonlinear Unmixing of Hyperspectral Images: Models and Algorithms. IEEE Signal Processing Magazine, 31, 82-94.
https://doi.org/10.1109/msp.2013.2279274
[15] Han, Z., Yang, J., Gao, L.R., et al. (2024) Dual-Branch Subpixel-Guided Network for Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing, 62, 1-13.
https://doi.org/10.1109/tgrs.2024.3418583
[16] Lin, T., Goyal, P., Girshick, R., He, K. and Dollar, P. (2017) Focal Loss for Dense Object Detection. 2017 IEEE International Conference on Computer Vision (ICCV), Venice, 22-29 October 2017, 2980-2988.
https://doi.org/10.1109/iccv.2017.324
[17] Zhao, L., Luo, W., Liao, Q., Chen, S. and Wu, J. (2022) Hyperspectral Image Classification with Contrastive Self-Supervised Learning under Limited Labeled Samples. IEEE Geoscience and Remote Sensing Letters, 19, 1-5.
https://doi.org/10.1109/lgrs.2022.3159549
[18] Wu, Q., He, M., Liu, Z. and Liu, Y. (2024) Multi-Scale Spatial-Spectral Residual Attention Network for Hyperspectral Image Classification. Electronics, 13, Article 262.
https://doi.org/10.3390/electronics13020262
[19] Chen, J., Li, W. and El-Askary, H. (2025) Spectral-Spatial Transformer with Multiscale Convolutional Attention for Hyperspectral Image Classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 18, 24223-24241.
https://doi.org/10.1109/jstars.2025.3608699
[20] Xu, Y., Zhang, L., Du, B. and Zhang, F. (2018) Spectral-Spatial Unified Networks for Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing, 56, 5893-5909.
https://doi.org/10.1109/tgrs.2018.2827407
[21] Meng, Z., Yan, Q., Zhao, F. and Liang, M. (2023) Hyperspectral Image Classification with Dynamic Spatial-Spectral Attention Network. 2023 13th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing (WHISPERS), Athens, 31 October 31-2 November 2023, 1-4.
https://doi.org/10.1109/whispers61460.2023.10431375
[22] Shu, Z., Wang, Y. and Yu, Z. (2024) Dual Attention Transformer Network for Hyperspectral Image Classification. Engineering Applications of Artificial Intelligence, 127, Article ID: 107351.
https://doi.org/10.1016/j.engappai.2023.107351
[23] Loshchilov, I. and Hutter, F. (2019) Decoupled Weight Decay Regularization. arXiv: 1711.05101
[24] Kingma, D.P. and Ba, J. (2015) Adam: A Method for Stochastic Optimization. arXiv: 1412.6980
https://arxiv.org/abs/1412.6980
[25] Ioffe, S. and Szegedy, C. (2015) Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. 32nd International Conference on Machine Learning, Lille, 6-11 July 2015, 448-456.
[26] Chen, L., Wei, Z. and Xu, Y. (2020) A Lightweight Spectral-Spatial Feature Extraction and Fusion Network for Hyperspectral Image Classification. Remote Sensing, 12, Article 1395.
https://doi.org/10.3390/rs12091395