Hyperspectral-Multispectral Image Fusion via Tensor Ring and Subspace Decompositions
Fusion from a spatially low resolution hyperspectral image (LR-HSI) and a spectrally low resolution multispectral image (MSI) to produce a high spatial-spectral HSI (HR-HSI), known as hyperspectral super resolution, has risen to a preferred topic for reinforcing the spatial-spectral resolution of HS...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9525254/ |
| _version_ | 1819112048747872256 |
|---|---|
| author | Honghui Xu Mengjie Qin Sheng yong Chen Yuhui Zheng Jian wei Zheng |
| author_facet | Honghui Xu Mengjie Qin Sheng yong Chen Yuhui Zheng Jian wei Zheng |
| author_sort | Honghui Xu |
| collection | DOAJ |
| description | Fusion from a spatially low resolution hyperspectral image (LR-HSI) and a spectrally low resolution multispectral image (MSI) to produce a high spatial-spectral HSI (HR-HSI), known as hyperspectral super resolution, has risen to a preferred topic for reinforcing the spatial-spectral resolution of HSI in recent years. In this work, we propose a new model, namely, low-rank tensor ring decomposition based on tensor nuclear norm (LRTRTNN), for HSI-MSI fusion. Specifically, for each spectrally subspace cube, similar patches are grouped to exploit both the global low-rank property of LR-HSI and the nonlocal similarity of HR-MSI. Afterward, a joint optimization of all groups via the presented LRTRTNN approximation is implemented in a unified cost function. With the introduced tensor nuclear norm (TNN) constraint, all 3D tensor ring factors are no longer unfolded to suit the matrix nuclear norm used in conventional methods, and the internal tensor structure can be naturally retained. The alternating direction method of multipliers is introduced for coefficients update. Numerical and visual experiments on real data show that our LRTRTNN method outperforms most state-of-the-art algorithms in terms of fusing performance. |
| first_indexed | 2024-12-22T04:07:19Z |
| format | Article |
| id | doaj.art-864e5a6127624bf2a7a929120f8e72ca |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-12-22T04:07:19Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-864e5a6127624bf2a7a929120f8e72ca2022-12-21T18:39:37ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352021-01-01148823883710.1109/JSTARS.2021.31082339525254Hyperspectral-Multispectral Image Fusion via Tensor Ring and Subspace DecompositionsHonghui Xu0https://orcid.org/0000-0002-6213-2979Mengjie Qin1https://orcid.org/0000-0001-6350-2217Sheng yong Chen2https://orcid.org/0000-0002-6705-3831Yuhui Zheng3https://orcid.org/0000-0002-1709-3093Jian wei Zheng4https://orcid.org/0000-0001-6017-0552College of Computer Science, and Technology, Zhejiang University of Technology, Hangzhou, ChinaCollege of Computer Science, and Technology, Zhejiang University of Technology, Hangzhou, ChinaCollege of Computer Sciences and Engineering, Tianjin University of Technology, Tianjin, ChinaSchool of Computer and Software, Nanjing University of Information Science and Technology, Nanjing, ChinaCollege of Computer Science, and Technology, Zhejiang University of Technology, Hangzhou, ChinaFusion from a spatially low resolution hyperspectral image (LR-HSI) and a spectrally low resolution multispectral image (MSI) to produce a high spatial-spectral HSI (HR-HSI), known as hyperspectral super resolution, has risen to a preferred topic for reinforcing the spatial-spectral resolution of HSI in recent years. In this work, we propose a new model, namely, low-rank tensor ring decomposition based on tensor nuclear norm (LRTRTNN), for HSI-MSI fusion. Specifically, for each spectrally subspace cube, similar patches are grouped to exploit both the global low-rank property of LR-HSI and the nonlocal similarity of HR-MSI. Afterward, a joint optimization of all groups via the presented LRTRTNN approximation is implemented in a unified cost function. With the introduced tensor nuclear norm (TNN) constraint, all 3D tensor ring factors are no longer unfolded to suit the matrix nuclear norm used in conventional methods, and the internal tensor structure can be naturally retained. The alternating direction method of multipliers is introduced for coefficients update. Numerical and visual experiments on real data show that our LRTRTNN method outperforms most state-of-the-art algorithms in terms of fusing performance.https://ieeexplore.ieee.org/document/9525254/Hyperspectral imaginghyperspectral super resolutionimage fusionlow-rank decompositionmultispectral image (MSI)tenson ring |
| spellingShingle | Honghui Xu Mengjie Qin Sheng yong Chen Yuhui Zheng Jian wei Zheng Hyperspectral-Multispectral Image Fusion via Tensor Ring and Subspace Decompositions IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Hyperspectral imaging hyperspectral super resolution image fusion low-rank decomposition multispectral image (MSI) tenson ring |
| title | Hyperspectral-Multispectral Image Fusion via Tensor Ring and Subspace Decompositions |
| title_full | Hyperspectral-Multispectral Image Fusion via Tensor Ring and Subspace Decompositions |
| title_fullStr | Hyperspectral-Multispectral Image Fusion via Tensor Ring and Subspace Decompositions |
| title_full_unstemmed | Hyperspectral-Multispectral Image Fusion via Tensor Ring and Subspace Decompositions |
| title_short | Hyperspectral-Multispectral Image Fusion via Tensor Ring and Subspace Decompositions |
| title_sort | hyperspectral multispectral image fusion via tensor ring and subspace decompositions |
| topic | Hyperspectral imaging hyperspectral super resolution image fusion low-rank decomposition multispectral image (MSI) tenson ring |
| url | https://ieeexplore.ieee.org/document/9525254/ |
| work_keys_str_mv | AT honghuixu hyperspectralmultispectralimagefusionviatensorringandsubspacedecompositions AT mengjieqin hyperspectralmultispectralimagefusionviatensorringandsubspacedecompositions AT shengyongchen hyperspectralmultispectralimagefusionviatensorringandsubspacedecompositions AT yuhuizheng hyperspectralmultispectralimagefusionviatensorringandsubspacedecompositions AT jianweizheng hyperspectralmultispectralimagefusionviatensorringandsubspacedecompositions |