volume | PIER Journals

Abstract

Although Transformer models have made significant progress in the field of hyperspectral image denoising, their original architecture still has limitations in processing the spatial and spectral correlations of images. It often results in the loss of details in spatial features and insufficient exploration of the uniqueness of different spectral bands. To overcome these challenges, this paper proposes a Transformer based spatial spectral attention network aimed at enhancing the utilization efficiency of spatial spectral correlations. In response to the common problem of over smoothing in spatial feature processing, a dual channel spatial feature fusion module is introduced, which effectively enhances the capture of spatial details and ensures clear reproduction of image textures and edges. Meanwhile, in the spectral dimension, a multi-scale spectral feature extraction with self-attention mechanism is applied, which can sensitively identify and utilize the differences between spectral bands, thereby achieving more accurate feature extraction at the spectral level. By integrating residual connections in the spatial spectral feature extraction layer, the model can efficiently fuse spatial and spectral information, ultimately achieving high-quality denoising. The experimental results have verified the excellent performance of this method on both the ICVL dataset and the Urban real dataset, achieving good denoising results and demonstrating significant advantages in maintaining image details and spectral fidelity.

1. Qin, Xuexiang, Yuxiang Zhang, and Yanni Dong, "Domain alignment dynamic spectral and spatial feature fusion for hyperspectral change detection," IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol. 18, 557-568, 2025.
doi:10.1109/jstars.2024.3495217

2. Song, Shao-Zhong, Yuan-Yuan Liu, Zi-Yang Zhou, Xing Teng, Ji-Hong Li, Jun-Ling Liu, and Xun Gao, "Identification of sorghum breed by hyperspectral image technology," Spectroscopy and Spectral Analysis, Vol. 44, No. 5, 1392-1397, 2024.
doi:10.3964/j.issn.1000-0593(2024)05-1392-06

3. Lu, Bing, Phuong D. Dao, Jiangui Liu, Yuhong He, and Jiali Shang, "Recent advances of hyperspectral imaging technology and applications in agriculture," Remote Sensing, Vol. 12, No. 16, 2659, 2020.
doi:10.3390/rs12162659

4. Zhang, Jianan, Qing Zhang, Jiansheng Wang, Yan Wang, and Qingli Li, "A dual branch based stitching method for whole slide hyperspectral pathological imaging," Displays, Vol. 89, 103090, 2025.
doi:10.1016/j.displa.2025.103090

5. Wu, Guoyong, Mohammed A. A. Al-Qaness, Dalal Al-Alimi, Abdelghani Dahou, Mohamed Abd Elaziz, and Ahmed A. Ewees, "Hyperspectral image classification using graph convolutional network: A comprehensive review," Expert Systems with Applications, Vol. 257, 125106, 2024.
doi:10.1016/j.eswa.2024.125106

6. Zhao, Xi-Le, Hao Zhang, Tai-Xiang Jiang, Michael K. Ng, and Xiong-Jun Zhang, "Fast algorithm with theoretical guarantees for constrained low-tubal-rank tensor recovery in hyperspectral images denoising," Neurocomputing, Vol. 413, 397-409, 2020.
doi:10.1016/j.neucom.2020.07.022

7. Zhang, Aiyi, Fulai Liu, and Ruiyan Du, "Probability-weighted tensor robust PCA with CP decomposition for hyperspectral image restoration," Signal Processing, Vol. 209, 109051, 2023.
doi:10.1016/j.sigpro.2023.109051

8. Zhang, Qiang, Yushuai Dong, Qiangqiang Yuan, Meiping Song, and Haoyang Yu, "Combined deep priors with low-rank tensor factorization for hyperspectral image restoration," IEEE Geoscience and Remote Sensing Letters, Vol. 20, 1-5, 2023.
doi:10.1109/lgrs.2023.3236341

9. Wang, Zhicheng, Michael K. Ng, Lina Zhuang, Lianru Gao, and Bing Zhang, "Nonlocal self-similarity-based hyperspectral remote sensing image denoising with 3-D convolutional neural network," IEEE Transactions on Geoscience and Remote Sensing, Vol. 60, 1-17, 2022.
doi:10.1109/tgrs.2022.3182144

10. Zha, Zhiyuan, Bihan Wen, Xin Yuan, Jiachao Zhang, Jiantao Zhou, Yilong Lu, and Ce Zhu, "Nonlocal structured sparsity regularization modeling for hyperspectral image denoising," IEEE Transactions on Geoscience and Remote Sensing, Vol. 61, 1-16, 2023.
doi:10.1109/tgrs.2023.3269224

11. Wang, Peng, Liguo Wang, Henry Leung, and Gong Zhang, "Super-resolution mapping based on spatial-spectral correlation for spectral imagery," IEEE Transactions on Geoscience and Remote Sensing, Vol. 59, No. 3, 2256-2268, 2020.
doi:10.1109/tgrs.2020.3004353

12. Zhou, Yang, Yong Chen, Jinshan Zeng, Wei He, and Min Huang, "Unidirectional spatial and spectral smoothed tensor ring decomposition for hyperspectral image denoising and destriping," IEEE Geoscience and Remote Sensing Letters, Vol. 21, 1-5, 2024.
doi:10.1109/lgrs.2024.3412804

13. Zhao, Shuheng, Xiaolin Zhu, Denghong Liu, Fei Xu, Yan Wang, Liupeng Lin, Xuehong Chen, and Qiangqiang Yuan, "A hyperspectral image denoising method based on land cover spectral autocorrelation," International Journal of Applied Earth Observation and Geoinformation, Vol. 123, 103481, 2023.
doi:10.1016/j.jag.2023.103481

14. Aharon, M., M. Elad, and A. Bruckstein, "K-SVD: An algorithm for designing overcomplete dictionaries for sparse representation," IEEE Transactions on Signal Processing, Vol. 54, No. 11, 4311-4322, 2006.
doi:10.1109/tsp.2006.881199

15. Dabov, Kostadin, Alessandro Foi, Vladimir Katkovnik, and Karen Egiazarian, "Image denoising by sparse 3-D transform-domain collaborative filtering," IEEE Transactions on Image Processing, Vol. 16, No. 8, 2080-2095, 2007.
doi:10.1109/tip.2007.901238

16. Chang, Yi, Luxin Yan, and Sheng Zhong, "Hyper-Laplacian regularized unidirectional low-rank tensor recovery for multispectral image denoising," 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 5901-5909, Honolulu, HI, USA, 2017.
doi:10.1109/cvpr.2017.625

17. Peng, Yi, Deyu Meng, Zongben Xu, Chenqiang Gao, Yi Yang, and Biao Zhang, "Decomposable nonlocal tensor dictionary learning for multispectral image denoising," 2014 IEEE Conference on Computer Vision and Pattern Recognition, 2949-2956, Columbus, OH, USA, 2014.
doi:10.1109/cvpr.2014.377

18. Peng, Jiangtao, Weiwei Sun, Heng-Chao Li, Wei Li, Xiangchao Meng, Chiru Ge, and Qian Du, "Low-rank and sparse representation for hyperspectral image processing: A review," IEEE Geoscience and Remote Sensing Magazine, Vol. 10, No. 1, 10-43, 2022.
doi:10.1109/mgrs.2021.3075491

19. Maggioni, Matteo, Vladimir Katkovnik, Karen Egiazarian, and Alessandro Foi, "Nonlocal transform-domain filter for volumetric data denoising and reconstruction," IEEE Transactions on Image Processing, Vol. 22, No. 1, 119-133, 2013.
doi:10.1109/tip.2012.2210725

20. Meng, Pinchao, Zhaobin Xu, Xianchao Wang, Weishi Yin, and Hongyu Liu, "A novel method for solving the inverse spectral problem with incomplete data," Journal of Computational and Applied Mathematics, Vol. 463, 116525, 2025.
doi:10.1016/j.cam.2025.116525

21. Jiang, Yan, Hongyu Liu, Tianhao Ni, and Kai Zhang, "Inverse problems for nonlinear progressive waves," Calculus of Variations and Partial Differential Equations, Vol. 64, No. 4, 116, 2025.
doi:10.1007/s00526-025-02964-0

22. Yin, Weishi, Zhengxuan Shen, Pinchao Meng, and Hongyu Liu, "An online interactive physics-informed adversarial network for solving mean field games," Engineering Analysis with Boundary Elements, Vol. 169, 106002, 2024.
doi:10.1016/j.enganabound.2024.106002

23. Nguyen, Han V., Magnus O. Ulfarsson, Jakob Sigurdsson, and Johannes R. Sveinsson, "Deep sparse and low-rank prior for hyperspectral image denoising," IGARSS 2022 --- 2022 IEEE International Geoscience and Remote Sensing Symposium, 1217-1220, Kuala Lumpur, Malaysia, 2022.
doi:10.1109/IGARSS46834.2022.9884071

24. Yuan, Qiangqiang, Qiang Zhang, Jie Li, Huanfeng Shen, and Liangpei Zhang, "Hyperspectral image denoising employing a spatial-spectral deep residual convolutional neural network," IEEE Transactions on Geoscience and Remote Sensing, Vol. 57, No. 2, 1205-1218, 2019.
doi:10.1109/tgrs.2018.2865197

25. Dusmanu, Mihai, Ignacio Rocco, Tomas Pajdla, Marc Pollefeys, Josef Sivic, Akihiko Torii, and Torsten Sattler, "D2-Net: A trainable CNN for joint description and detection of local features," 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 8084-8093, Long Beach, CA, USA, 2019.
doi:10.1109/cvpr.2019.00828

26. Wei, Kaixuan, Ying Fu, and Hua Huang, "3-D quasi-recurrent neural network for hyperspectral image denoising," IEEE Transactions on Neural Networks and Learning Systems, Vol. 32, No. 1, 363-375, 2021.
doi:10.1109/tnnls.2020.2978756

27. Maffei, Alessandro, Juan M. Haut, Mercedes Eugenia Paoletti, Javier Plaza, Lorenzo Bruzzone, and Antonio Plaza, "A single model CNN for hyperspectral image denoising," IEEE Transactions on Geoscience and Remote Sensing, Vol. 58, No. 4, 2516-2529, 2020.
doi:10.1109/tgrs.2019.2952062

28. Wang, Wenhai, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, and Ling Shao, "Pyramid vision transformer: A versatile backbone for dense prediction without convolutions," 2021 IEEE/CVF International Conference on Computer Vision (ICCV), 568-578, Montreal, QC, Canada, 2021.
doi:10.1109/ICCV48922.2021.00061

29. Li, Miaoyu, Ji Liu, Ying Fu, Yulun Zhang, and Dejing Dou, "Spectral enhanced rectangle transformer for hyperspectral image denoising," 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 5805-5814, Vancouver, BC, Canada, 2023.
doi:10.1109/CVPR52729.2023.00562

30. Zhang, Qiang, Yushuai Dong, Yaming Zheng, Haoyang Yu, Meiping Song, Lifu Zhang, and Qiangqiang Yuan, "Three-dimension spatial-spectral attention transformer for hyperspectral image denoising," IEEE Transactions on Geoscience and Remote Sensing, Vol. 62, 1-13, 2024.
doi:10.1109/tgrs.2024.3458174

31. Wang, Zhendong, Xiaodong Cun, Jianmin Bao, Wengang Zhou, Jianzhuang Liu, and Houqiang Li, "Uformer: A general U-shaped transformer for image restoration," 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 17683-17693, New Orleans, LA, USA, 2022.
doi:10.1109/CVPR52688.2022.01716

32. Pang, Li, Weizhen Gu, and Xiangyong Cao, "TRQ3DNet: A 3D quasi-recurrent and transformer based network for hyperspectral image denoising," Remote Sensing, Vol. 14, No. 18, 4598, 2022.
doi:10.3390/rs14184598

33. Dosovitskiy, Alexey, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby, "An image is worth 16 × 16 words: Transformers for image recognition at scale," arXiv:2010.11929, 2020.
doi:https://doi.org/10.48550/arXiv.2010.11929

34. Arad, Boaz and Ohad Ben-Shahar, "Sparse recovery of hyperspectral signal from natural RGB images," Computer Vision-ECCV 2016: 14th European Conference, 19-34, Amsterdam, The Netherlands, 2016.
doi:10.1007/978-3-319-46478-7_2

35. Kalman, Linda S. and Edward M. Bassett III, "Classification and material identification in an urban environment using HYDICE hyperspectral data," Imaging Spectrometry III, Vol. 3118, 57-68, 1997.
doi:10.1117/12.283843

36. Kingma, Diederik P. and Jimmy Ba, "Adam: A method for stochastic optimization," The 3rd International Conference for Learning Representations, San Diego, USA, 2015.
doi:https://doi.org/10.48550/arXiv.1412.6980

Dr. Xiao-Zhen Ren

Dr. Jing Cui

Dr. Yi Hu

Dr. Zhipeng Guo

Dr. Yingying Niu