Hyperspectral Image Denoising via Low-Rank Representation and CNN Denoiser
Hyperspectral images (HSIs) are widely used in various tasks such as earth observation and target detection. However, during the imaging process, HSIs are often corrupted by various noises. In this article, we firstly investigate the advantages of traditional physical restoration models and the deno...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2022-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9664348/ |
| _version_ | 1828147503313190912 |
|---|---|
| author | Hezhi Sun Ming Liu Ke Zheng Dong Yang Jindong Li Lianru Gao |
| author_facet | Hezhi Sun Ming Liu Ke Zheng Dong Yang Jindong Li Lianru Gao |
| author_sort | Hezhi Sun |
| collection | DOAJ |
| description | Hyperspectral images (HSIs) are widely used in various tasks such as earth observation and target detection. However, during the imaging process, HSIs are often corrupted by various noises. In this article, we firstly investigate the advantages of traditional physical restoration models and the denoising convolutional neural networks (CNN) for HSIs denoising tasks. The sparse based low-rank representation can explore the global correlations in both the spatial and spectral domains, and the CNN-based denoiser can represent the deep prior which cannot be designed by traditional restoration models. Then, we propose a HSI denoising model with low-rank representation and CNN denoiser prior in the flexible and extensible plug-and-play framework by combining the advantages of the two methods. The proposed model is user-friendly, requiring no retraining. Simulated data experiments show that, compared with competitive methods, the proposed one achieves better denoising results for both additive Gaussian noise and Poissonian noise in various quantitative evaluation indicators. Real data experiments show that the proposed model yields the best performance. |
| first_indexed | 2024-04-11T21:02:22Z |
| format | Article |
| id | doaj.art-d5a288cc7a324e138741c98d131f8077 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-04-11T21:02:22Z |
| publishDate | 2022-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-d5a288cc7a324e138741c98d131f80772022-12-22T04:03:28ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352022-01-011571672810.1109/JSTARS.2021.31385649664348Hyperspectral Image Denoising via Low-Rank Representation and CNN DenoiserHezhi Sun0https://orcid.org/0000-0001-6876-6051Ming Liu1https://orcid.org/0000-0002-1846-6445Ke Zheng2https://orcid.org/0000-0002-0108-0511Dong Yang3Jindong Li4Lianru Gao5https://orcid.org/0000-0003-3888-8124Research Center of Satellite Technology, Harbin Institute of Technology, Harbin, ChinaResearch Center of Satellite Technology, Harbin Institute of Technology, Harbin, ChinaKey Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaChinese Academy of Sciences, Beijing, ChinaChinese Academy of Sciences, Beijing, ChinaKey Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaHyperspectral images (HSIs) are widely used in various tasks such as earth observation and target detection. However, during the imaging process, HSIs are often corrupted by various noises. In this article, we firstly investigate the advantages of traditional physical restoration models and the denoising convolutional neural networks (CNN) for HSIs denoising tasks. The sparse based low-rank representation can explore the global correlations in both the spatial and spectral domains, and the CNN-based denoiser can represent the deep prior which cannot be designed by traditional restoration models. Then, we propose a HSI denoising model with low-rank representation and CNN denoiser prior in the flexible and extensible plug-and-play framework by combining the advantages of the two methods. The proposed model is user-friendly, requiring no retraining. Simulated data experiments show that, compared with competitive methods, the proposed one achieves better denoising results for both additive Gaussian noise and Poissonian noise in various quantitative evaluation indicators. Real data experiments show that the proposed model yields the best performance.https://ieeexplore.ieee.org/document/9664348/Convolutional neural network (CNN)hyperspectral image (HSI) denoisinglow-rank representation |
| spellingShingle | Hezhi Sun Ming Liu Ke Zheng Dong Yang Jindong Li Lianru Gao Hyperspectral Image Denoising via Low-Rank Representation and CNN Denoiser IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Convolutional neural network (CNN) hyperspectral image (HSI) denoising low-rank representation |
| title | Hyperspectral Image Denoising via Low-Rank Representation and CNN Denoiser |
| title_full | Hyperspectral Image Denoising via Low-Rank Representation and CNN Denoiser |
| title_fullStr | Hyperspectral Image Denoising via Low-Rank Representation and CNN Denoiser |
| title_full_unstemmed | Hyperspectral Image Denoising via Low-Rank Representation and CNN Denoiser |
| title_short | Hyperspectral Image Denoising via Low-Rank Representation and CNN Denoiser |
| title_sort | hyperspectral image denoising via low rank representation and cnn denoiser |
| topic | Convolutional neural network (CNN) hyperspectral image (HSI) denoising low-rank representation |
| url | https://ieeexplore.ieee.org/document/9664348/ |
| work_keys_str_mv | AT hezhisun hyperspectralimagedenoisingvialowrankrepresentationandcnndenoiser AT mingliu hyperspectralimagedenoisingvialowrankrepresentationandcnndenoiser AT kezheng hyperspectralimagedenoisingvialowrankrepresentationandcnndenoiser AT dongyang hyperspectralimagedenoisingvialowrankrepresentationandcnndenoiser AT jindongli hyperspectralimagedenoisingvialowrankrepresentationandcnndenoiser AT lianrugao hyperspectralimagedenoisingvialowrankrepresentationandcnndenoiser |