Abstract
A lot of hyperspectral images (HSIs) are corrupted by noises when they are captured. Noise removal is an essential pre-processing for the noisy HSIs. Though denoising algorithms for common (grayscale or RGB) images have been studied for decades, HSIs have their inherent characteristics, so denoising algorithms for HSIs need to be specially designed. In this work, we have developed a non-local denoising algorithm for HSIs based on multi-task graph-regularized sparse nonnegative matrix factorization (MTGSNMF). MTGSNMF delivers noise removal in both spatial and spectral views. In spatial view, patch-based sparse recovery is performed by sparse nonnegative matrix factorization (SNMF), which conducts noise suppression and local pattern preservation. Graph regularization is imposed on the SNMF model for maintaining the non-local similarities between patches. In spectral view, spectral structure is extracted by multi-task learning, i.e., denoising tasks of different bands are bound by sharing the same coefficient matrix. By exploiting the non-local similarity in spatial view and spectral structure in spectral view, MTGSNMF achieves superior denoising performance on HSI datasets.
Supported partly by the National Natural Science Foundation of China (grant numbers 61701468 and 62071421), and partly by the National Key Research and Development Program of China (grant number 2018YFB0505000).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The code of BM4D is available at https://www.cs.tut.fi/~foi/GCF-BM3D/BM4D_v3p2.zip.
- 2.
The code of NLMF is available at https://www.mathworks.com/matlabcentral/fileexchange/27395-fast-non-local-means-1d-2d-color-and-3d.
- 3.
The code of NGmeet is available at https://github.com/quanmingyao/NGMeet.
- 4.
The code of the proposed MTGSNMF is available at https://github.com/yeminchao/MTGSNMF.
References
Aggarwal, H.K., Majumdar, A.: Hyperspectral image denoising using spatio-spectral total variation. IEEE Geosci. Remote Sens. Lett. 13(3), 442–446 (2016)
Bioucas-Dias, J.M., Nascimento, J.M.P.: Hyperspectral subspace identification. IEEE Trans. Geosci. Remote Sens. 46(8), 2435–2445 (2008)
Buades, A., Coll, B., Morel, J.: A non-local algorithm for image denoising. In: Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition, vol. 2, pp. 60–65 (2005)
Cai, D., He, X., Han, J., Huang, T.S.: Graph regularized nonnegative matrix factorization for data representation. IEEE Trans. Pattern Anal. Mach. Intell. 33(8), 1548–1560 (2011)
Camps-Valls, G., Bruzzone, L.: Kernel-based methods for hyperspectral image classification. IEEE Trans. Geosci. Remote Sens. 43(6), 1351–1362 (2005)
Chen, H., Ye, M., Lu, H., Lei, L., Qian, Y.: Dual dictionary learning for mining a unified feature subspace between different hyperspectral image scenes. In: Proceedings of IEEE International Geoscience and Remote Sensing Symposium, pp. 1096–1099 (2019)
Dabov, K., Foi, A., Katkovnik, V., Egiazarian, K.: Image denoising by sparse 3-D transform-domain collaborative filtering. IEEE Trans. Image Process. 16(8), 2080–2095 (2007)
Dabov, K., Foi, A., Katkovnik, V., Egiazarian, K.: Image denoising with block-matching and 3D filtering. In: Proceedings of SPIE, Image Processing: Algorithms and Systems, Neural Networks, and Machine Learning, pp. 354–365 (2006)
Dian, R., Fang, L., Li, S.: Hyperspectral image super-resolution via non-local sparse tensor factorization. In: Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition (2017)
He, W., Yao, Q., Li, C., Yokoya, N., Zhao, Q.: Non-local meets global: an integrated paradigm for hyperspectral denoising. In: Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 6861–6870 (2019)
Jia, S., Ji, Z., Qian, Y., Shen, L.: Unsupervised band selection for hyperspectral imagery classification without manual band removal. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 5(2), 531–543 (2012)
Li, J., Yuan, Q., Shen, H., Zhang, L.: Hyperspectral image recovery employing a multidimensional nonlocal total variation model. Signal Process 111, 230–248 (2015)
Maggioni, M., Katkovnik, V., Egiazarian, K., Foi, A.: Nonlocal transform-domain filter for volumetric data denoising and reconstruction. IEEE Trans. Image Process. 22(1), 119–133 (2013)
Mairal, J., Bach, F., Ponce, J., Sapiro, G., Zisserman, A.: Non-local sparse models for image restoration. In: Proceedings of IEEE International Conference On Computer Vision, pp. 2272–2279 (2009)
Qian, Y., Shen, Y., Ye, M., Wang, Q.: 3-D nonlocal means filter with noise estimation for hyperspectral imagery denoising. In: Proceedings of IEEE International Geoscience and Remote Sensing Symposium, pp. 1345–1348 (2012)
Qian, Y., Ye, M.: Hyperspectral imagery restoration using nonlocal spectral-spatial structured sparse representation with noise estimation. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 6(2), 499–515 (2013)
Xiong, F., Zhou, J., Qian, Y.: Hyperspectral restoration via \(l_0\) gradient regularized low-rank tensor factorization. IEEE Trans. Geosci. Remote Sens. 57(12), 10410–10425 (2019)
Xu, P., Chen, B., Xue, L., Zhang, J., Zhu, L., Duan, H.: A new MNF-BM4D denoising algorithm based on guided filtering for hyperspectral images. ISA Trans. 92, 315–324 (2019)
Ye, M., Qian, Y., Zhou, J.: Multitask sparse nonnegative matrix factorization for joint spectral-spatial hyperspectral imagery denoising. IEEE Trans. Geosci. Remote Sens. 53(5), 2621–2639 (2015)
Ye, M., Chen, H., Ji, C., Lei, L., Qian, Y.: Spectral-spatial joint noise estimation for hyperspectral images. In: Proceedings of International Geoscience and Remote Sensing Symposium, pp. 230–233 (2019)
Ye, M., Zheng, W., Lu, H., Zeng, X., Qian, Y.: Cross-scene hyperspectral image classification based on DWT and manifold-constrained subspace learning. Int. J. Wavelets Multiresolut. Inf. Process. 15(06), 1750062 (2017)
Zhang, H., Li, J., Huang, Y., Zhang, L.: A nonlocal weighted joint sparse representation classification method for hyperspectral imagery. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 7(6), 2056–2065 (2014)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Lei, L., Huang, B., Ye, M., Chen, H., Qian, Y. (2021). A Graph-Regularized Non-local Hyperspectral Image Denoising Method. In: Nguyen, M., Yan, W.Q., Ho, H. (eds) Geometry and Vision. ISGV 2021. Communications in Computer and Information Science, vol 1386. Springer, Cham. https://doi.org/10.1007/978-3-030-72073-5_25
Download citation
DOI: https://doi.org/10.1007/978-3-030-72073-5_25
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-72072-8
Online ISBN: 978-3-030-72073-5
eBook Packages: Computer ScienceComputer Science (R0)