Hyperspectral Image Mixed Noise Removal Using a Subspace Projection Attention and Residual Channel Attention Network
Although the existing deep-learning-based hyperspectral image (HSI) denoising methods have achieved tremendous success, recovering high-quality HSIs in complex scenes that contain mixed noise is still challenging. Besides, these methods have not fully explored the local and global spatial–spectral i...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-04-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/9/2071 |
| _version_ | 1827671664837525504 |
|---|---|
| author | Hezhi Sun Ke Zheng Ming Liu Chao Li Dong Yang Jindong Li |
| author_facet | Hezhi Sun Ke Zheng Ming Liu Chao Li Dong Yang Jindong Li |
| author_sort | Hezhi Sun |
| collection | DOAJ |
| description | Although the existing deep-learning-based hyperspectral image (HSI) denoising methods have achieved tremendous success, recovering high-quality HSIs in complex scenes that contain mixed noise is still challenging. Besides, these methods have not fully explored the local and global spatial–spectral information of HSIs. To address the above issues, a novel HSI mixed noise removal network called subspace projection attention and residual channel attention network (SPARCA-Net) is proposed. Specifically, we propose an orthogonal subspace projection attention (OSPA) module to adaptively learn to generate bases of the signal subspace and project the input into such space to remove noise. By leveraging the local and global spatial relations, OSPA is able to reconstruct the local structure of the feature maps more precisely. We further propose a residual channel attention (RCA) module to emphasize the interdependence between feature maps and exploit the global channel correlation of them, which could enhance the channel-wise adaptive learning. In addition, multiscale joint spatial–spectral input and residual learning strategies are employed to capture multiscale spatial–spectral features and reduce the degradation problem, respectively. Synthetic and real HSI data experiments demonstrated that the proposed HSI denoising network outperforms many of the advanced methods in both quantitative and qualitative assessments. |
| first_indexed | 2024-03-10T03:45:03Z |
| format | Article |
| id | doaj.art-8eaa02b428ff432eb00ab4944a39ac61 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T03:45:03Z |
| publishDate | 2022-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-8eaa02b428ff432eb00ab4944a39ac612023-11-23T09:10:10ZengMDPI AGRemote Sensing2072-42922022-04-01149207110.3390/rs14092071Hyperspectral Image Mixed Noise Removal Using a Subspace Projection Attention and Residual Channel Attention NetworkHezhi Sun0Ke Zheng1Ming Liu2Chao Li3Dong Yang4Jindong Li5Research Center of Satellite Technology, Harbin Institute of Technology, Harbin 150001, ChinaKey Laboratory of Computational Optical Imaging Technology, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaResearch Center of Satellite Technology, Harbin Institute of Technology, Harbin 150001, ChinaResearch Center for Space Optical Engineering, Harbin Institute of Technology, Harbin 150001, ChinaChina Academy of Space Technology, Beijing 100094, ChinaChina Academy of Space Technology, Beijing 100094, ChinaAlthough the existing deep-learning-based hyperspectral image (HSI) denoising methods have achieved tremendous success, recovering high-quality HSIs in complex scenes that contain mixed noise is still challenging. Besides, these methods have not fully explored the local and global spatial–spectral information of HSIs. To address the above issues, a novel HSI mixed noise removal network called subspace projection attention and residual channel attention network (SPARCA-Net) is proposed. Specifically, we propose an orthogonal subspace projection attention (OSPA) module to adaptively learn to generate bases of the signal subspace and project the input into such space to remove noise. By leveraging the local and global spatial relations, OSPA is able to reconstruct the local structure of the feature maps more precisely. We further propose a residual channel attention (RCA) module to emphasize the interdependence between feature maps and exploit the global channel correlation of them, which could enhance the channel-wise adaptive learning. In addition, multiscale joint spatial–spectral input and residual learning strategies are employed to capture multiscale spatial–spectral features and reduce the degradation problem, respectively. Synthetic and real HSI data experiments demonstrated that the proposed HSI denoising network outperforms many of the advanced methods in both quantitative and qualitative assessments.https://www.mdpi.com/2072-4292/14/9/2071hyperspectral imagedenoisingattention networkdeep learning |
| spellingShingle | Hezhi Sun Ke Zheng Ming Liu Chao Li Dong Yang Jindong Li Hyperspectral Image Mixed Noise Removal Using a Subspace Projection Attention and Residual Channel Attention Network Remote Sensing hyperspectral image denoising attention network deep learning |
| title | Hyperspectral Image Mixed Noise Removal Using a Subspace Projection Attention and Residual Channel Attention Network |
| title_full | Hyperspectral Image Mixed Noise Removal Using a Subspace Projection Attention and Residual Channel Attention Network |
| title_fullStr | Hyperspectral Image Mixed Noise Removal Using a Subspace Projection Attention and Residual Channel Attention Network |
| title_full_unstemmed | Hyperspectral Image Mixed Noise Removal Using a Subspace Projection Attention and Residual Channel Attention Network |
| title_short | Hyperspectral Image Mixed Noise Removal Using a Subspace Projection Attention and Residual Channel Attention Network |
| title_sort | hyperspectral image mixed noise removal using a subspace projection attention and residual channel attention network |
| topic | hyperspectral image denoising attention network deep learning |
| url | https://www.mdpi.com/2072-4292/14/9/2071 |
| work_keys_str_mv | AT hezhisun hyperspectralimagemixednoiseremovalusingasubspaceprojectionattentionandresidualchannelattentionnetwork AT kezheng hyperspectralimagemixednoiseremovalusingasubspaceprojectionattentionandresidualchannelattentionnetwork AT mingliu hyperspectralimagemixednoiseremovalusingasubspaceprojectionattentionandresidualchannelattentionnetwork AT chaoli hyperspectralimagemixednoiseremovalusingasubspaceprojectionattentionandresidualchannelattentionnetwork AT dongyang hyperspectralimagemixednoiseremovalusingasubspaceprojectionattentionandresidualchannelattentionnetwork AT jindongli hyperspectralimagemixednoiseremovalusingasubspaceprojectionattentionandresidualchannelattentionnetwork |