噪声强度指导小波域的高光谱图像去噪
Noise Intensity Guided Wavelet Domain Denoising of Hyperspectral Images
DOI: 10.12677/mos.2025.145451, PDF,   
作者: 张人丹：上海理工大学光电信息与计算机工程学院，上海
关键词: 高光谱图像(HSI)；小波变换；窗口注意力；知识蒸馏；Hyperspectral Image (HSI)； Wavelet Transform； Window Attention； Knowledge Distillation
摘要: 如何在不破坏边缘细节结构的情况下去除各种噪声，是高光谱图像(HSI)要解决的问题。文章提出了一种噪声强度指导小波域网络，即NGWDNet，它应用学习得到的噪声特征图来指导小波变换后的频域特征，从而完全去除高光谱图像的噪声。首先，通过设计的噪声估计模块(NEB)精准计算出各波段的噪声特征图。随后利用离散小波变换(DWT)将原始HSI分离为两个高频分支与一个低频分支。针对不同频率分支，NGWDNet采用差异化处理策略。对于富含纹理和细节特征的高频分支，将噪声图与之融合，并设计基于移位/窗口的多头自注意力融合模块(W/SW-MSAF Block)，强化对细节信息的捕捉与处理。对于具有较多平滑结构特征的低频分支，则直接应用3D和2D Unit，高效保留结构信息。同时，充分考量不同分支间的特征交互，借助高频特征引导低频分支精细结构的重建，进一步挖掘纹理细节特征。最后，部署了空间–光谱残差块(SSRB)，在融合双域的同时进一步探索空间–光谱相关性，以获得更好的去噪输出。在公开数据集上进行了定性和定量实验，结果表明，NGWDNet在可视化和客观评价指标方面优于最先进的方法。
Abstract: How to remove all kinds of noise without destroying the edge detail structure is a problem to be solved in hyperspectral images (HSI). This paper proposes a noise intensity guided wavelet domain network, namely NGWDNet, which uses the learned noise feature map to guide the frequency domain features after wavelet transform so as to completely remove the noise of the hyperspectral image. Firstly, the noise characteristic map of each band is accurately calculated by the designed noise estimation module (NEB). Then, the original HSI is separated into two high-frequency branches and one low-frequency branch by discrete wavelet transform (DWT). For different frequency branches, NGWDNet adopts a differentiated processing strategy. For high-frequency branches rich in texture and detail features, the noise map is fused with it, and a multi-head self-attention fusion module based on shift/window (W/SW-MSAF Block) is designed to enhance the capture and processing of detail information. For low-frequency branches with smoother structural features, 3D and 2D Units are directly applied to efficiently retain structural information. At the same time, the feature interaction between different branches is fully considered, and the high-frequency features are used to guide the reconstruction of the fine structure of the low-frequency branches to further explore the texture detail features. Finally, we deployed the spatial-spectral residual block (SSRB) to further explore the spatial-spectral correlation while fusing the dual-domain to obtain better denoising output. We conducted qualitative and quantitative experiments on public datasets. The results show that NGWDNet is superior to the most advanced methods in terms of visualization and objective evaluation indicators.
文章引用：张人丹. 噪声强度指导小波域的高光谱图像去噪[J]. 建模与仿真, 2025, 14(5): 987-998. https://doi.org/10.12677/mos.2025.145451

参考文献

[1] Mohd Asaari, M.S., Mishra, P., Mertens, S., Dhondt, S., Inzé, D., Wuyts, N., et al. (2018) Close-Range Hyperspectral Image Analysis for the Early Detection of Stress Responses in Individual Plants in a High-Throughput Phenotyping Platform. ISPRS Journal of Photogrammetry and Remote Sensing, 138, 121-138. [Google Scholar] [CrossRef] 
[2] Capelle, V. and Hartmann, J.-M. (2022) Use of Hyperspectral Sounders to Retrieve Daytime Sea-Surface Temperature from Mid-Infrared Radiances: Application to Iasi. Remote Sensing of Environment, 280, Article 113171. [Google Scholar] [CrossRef] 
[3] Transon, J., D’Andrimont, R., Maugnard, A. and Defourny, P. (2018) Survey of Hyperspectral Earth Observation Applications from Space in the Sentinel-2 Context. Remote Sensing, 10, Article157. [Google Scholar] [CrossRef] 
[4] He, W., Chen, Y., Yokoya, N., Li, C. and Zhao, Q. (2022) Hyperspectral Super-Resolution via Coupled Tensor Ring Factorization. Pattern Recognition, 122, Article 108280. [Google Scholar] [CrossRef] 
[5] Xu, Y., Gong, J., Huang, X., Hu, X., Li, J., Li, Q., et al. (2022) Luojia-HSSR: A High Spatial-Spectral Resolution Remote Sensing Dataset for Land-Cover Classification with a New 3D-HRNet. Geo-Spatial Information Science, 26, 289-301. [Google Scholar] [CrossRef] 
[6] Qin, H., Xie, W., Li, Y., Jiang, K., Lei, J. and Du, Q. (2023) Weakly Supervised Adversarial Learning via Latent Space for Hyperspectral Target Detection. Pattern Recognition, 135, Article 109125. [Google Scholar] [CrossRef] 
[7] Wei, K., Fu, Y. and Huang, H. (2021) 3-D Quasi-Recurrent Neural Network for Hyperspectral Image Denoising. IEEE Transactions on Neural Networks and Learning Systems, 32, 363-375. [Google Scholar] [CrossRef] [PubMed]
[8] Huang, J. and Dragotti, P.L. (2022) WINNet: Wavelet-Inspired Invertible Network for Image Denoising. IEEE Transactions on Image Processing, 31, 4377-4392. [Google Scholar] [CrossRef] [PubMed]
[9] Liu, P., Zhang, H., Zhang, K., Lin, L. and Zuo, W. (2018) Multi-Level Wavelet-CNN for Image Restoration. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Salt Lake City, 18-22 June 2018, 773-782. [Google Scholar] [CrossRef] 
[10] Huang, H., He, R., Sun, Z. and Tan, T. (2017) Wavelet-SRNet: A Wavelet-Based CNN for Multi-Scale Face Super Resolution. 2017 IEEE International Conference on Computer Vision (ICCV), Venice, 22-29 October 2017, 1698-1706. [Google Scholar] [CrossRef] 
[11] Arad, B. and Ben-Shahar, O. (2016) Sparse Recovery of Hyperspectral Signal from Natural RGB Images. In: Leibe, B., Matas, J., Sebe, N. and Welling, M., Eds., Lecture Notes in Computer Science, Springer International Publishing, 19-34. [Google Scholar] [CrossRef] 
[12] Landgrebe, D.A. (2003) Signal Theory Methods in Multispectral Remote Sensing. Wiley, [Google Scholar] [CrossRef] 
[13] Mnih, V. and Hinton, G.E. (2010) Learning to Detect Roads in High-Resolution Aerial Images. In: Daniilidis, K., Maragos, P. and Paragios, N., Eds., Lecture Notes in Computer Science, Springer Berlin Heidelberg, 210-223. [Google Scholar] [CrossRef] 
[14] Maggioni, M., Katkovnik, V., Egiazarian, K. and Foi, A. (2013) Nonlocal Transform-Domain Filter for Volumetric Data Denoising and Reconstruction. IEEE Transactions on Image Processing, 22, 119-133. [Google Scholar] [CrossRef] [PubMed]
[15] Peng, Y., Meng, D., Xu, Z., Gao, C., Yang, Y. and Zhang, B. (2014) Decomposable Nonlocal Tensor Dictionary Learning for Multispectral Image Denoising. 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, 23-28 June 2014, 2949-2956. [Google Scholar] [CrossRef] 
[16] He, W., Zhang, H., Zhang, L. and Shen, H. (2016) Total-Variation-Regularized Low-Rank Matrix Factorization for Hyperspectral Image Restoration. IEEE Transactions on Geoscience and Remote Sensing, 54, 178-188. [Google Scholar] [CrossRef] 
[17] Wang, Y., Peng, J., Zhao, Q., Leung, Y., Zhao, X. and Meng, D. (2018) Hyperspectral Image Restoration via Total Variation Regularized Low-Rank Tensor Decomposition. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 11, 1227-1243. [Google Scholar] [CrossRef] 
[18] Yuan, Q., Zhang, Q., Li, J., Shen, H. and Zhang, L. (2019) Hyperspectral Image Denoising Employing a Spatial-Spectral Deep Residual Convolutional Neural Network. IEEE Transactions on Geoscience and Remote Sensing, 57, 1205-1218. [Google Scholar] [CrossRef] 
[19] Chang, Y., Yan, L., Fang, H., Zhong, S. and Liao, W. (2019) HSI-Denet: Hyperspectral Image Restoration via Convolutional Neural Network. IEEE Transactions on Geoscience and Remote Sensing, 57, 667-682. [Google Scholar] [CrossRef] 
[20] Pan, E., Ma, Y., Mei, X., Huang, J., Fan, F. and Ma, J. (2023) D2Net: Deep Denoising Network in Frequency Domain for Hyperspectral Image. IEEE/CAA Journal of Automatica Sinica, 10, 813-815. [Google Scholar] [CrossRef] 

为你推荐