A Constrained Convex Optimization Approach to Hyperspectral Image Restoration with Hybrid Spatio-Spectral Regularization
We propose a new constrained optimization approach to hyperspectral (HS) image restoration. Most existing methods restore a desirable HS image by solving some optimization problems, consisting of a regularization term(s) and a data-fidelity term(s). The methods have to handle a regularization term(s...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-10-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/12/21/3541 |
| _version_ | 1797549565954490368 |
|---|---|
| author | Saori Takeyama Shunsuke Ono Itsuo Kumazawa |
| author_facet | Saori Takeyama Shunsuke Ono Itsuo Kumazawa |
| author_sort | Saori Takeyama |
| collection | DOAJ |
| description | We propose a new constrained optimization approach to hyperspectral (HS) image restoration. Most existing methods restore a desirable HS image by solving some optimization problems, consisting of a regularization term(s) and a data-fidelity term(s). The methods have to handle a regularization term(s) and a data-fidelity term(s) simultaneously in one objective function; therefore, we need to carefully control the hyperparameter(s) that balances these terms. However, the setting of such hyperparameters is often a troublesome task because their suitable values depend strongly on the regularization terms adopted and the noise intensities on a given observation. Our proposed method is formulated as a convex optimization problem, utilizing a novel hybrid regularization technique named Hybrid Spatio-Spectral Total Variation (HSSTV) and incorporating data-fidelity as hard constraints. HSSTV has a strong noise and artifact removal ability while avoiding oversmoothing and spectral distortion, without combining other regularizations such as low-rank modeling-based ones. In addition, the constraint-type data-fidelity enables us to translate the hyperparameters that balance between regularization and data-fidelity to the upper bounds of the degree of data-fidelity that can be set in a much easier manner. We also develop an efficient algorithm based on the alternating direction method of multipliers (ADMM) to efficiently solve the optimization problem. We illustrate the advantages of the proposed method over various HS image restoration methods through comprehensive experiments, including state-of-the-art ones. |
| first_indexed | 2024-03-10T15:16:22Z |
| format | Article |
| id | doaj.art-f102f1d300324a43a68f1f80297d5528 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T15:16:22Z |
| publishDate | 2020-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-f102f1d300324a43a68f1f80297d55282023-11-20T18:56:26ZengMDPI AGRemote Sensing2072-42922020-10-011221354110.3390/rs12213541A Constrained Convex Optimization Approach to Hyperspectral Image Restoration with Hybrid Spatio-Spectral RegularizationSaori Takeyama0Shunsuke Ono1Itsuo Kumazawa2Department of Information and Communications Engineering, School of Engineering, Tokyo Institute of Technology, Kanagawa 226-8503, JapanDepartment of Computer Science, School of Computing, Tokyo Institute of Technology, Kanagawa 226-8503, JapanLaboratory for Future Interdisciplinary Research of Science and Technology (FIRST), Institute of Innovative Research (IIR), Tokyo Institute of Technology, Kanagawa 226-8503, JapanWe propose a new constrained optimization approach to hyperspectral (HS) image restoration. Most existing methods restore a desirable HS image by solving some optimization problems, consisting of a regularization term(s) and a data-fidelity term(s). The methods have to handle a regularization term(s) and a data-fidelity term(s) simultaneously in one objective function; therefore, we need to carefully control the hyperparameter(s) that balances these terms. However, the setting of such hyperparameters is often a troublesome task because their suitable values depend strongly on the regularization terms adopted and the noise intensities on a given observation. Our proposed method is formulated as a convex optimization problem, utilizing a novel hybrid regularization technique named Hybrid Spatio-Spectral Total Variation (HSSTV) and incorporating data-fidelity as hard constraints. HSSTV has a strong noise and artifact removal ability while avoiding oversmoothing and spectral distortion, without combining other regularizations such as low-rank modeling-based ones. In addition, the constraint-type data-fidelity enables us to translate the hyperparameters that balance between regularization and data-fidelity to the upper bounds of the degree of data-fidelity that can be set in a much easier manner. We also develop an efficient algorithm based on the alternating direction method of multipliers (ADMM) to efficiently solve the optimization problem. We illustrate the advantages of the proposed method over various HS image restoration methods through comprehensive experiments, including state-of-the-art ones.https://www.mdpi.com/2072-4292/12/21/3541hyperspectral image restorationADMMmixed noise removalcompressed sensing |
| spellingShingle | Saori Takeyama Shunsuke Ono Itsuo Kumazawa A Constrained Convex Optimization Approach to Hyperspectral Image Restoration with Hybrid Spatio-Spectral Regularization Remote Sensing hyperspectral image restoration ADMM mixed noise removal compressed sensing |
| title | A Constrained Convex Optimization Approach to Hyperspectral Image Restoration with Hybrid Spatio-Spectral Regularization |
| title_full | A Constrained Convex Optimization Approach to Hyperspectral Image Restoration with Hybrid Spatio-Spectral Regularization |
| title_fullStr | A Constrained Convex Optimization Approach to Hyperspectral Image Restoration with Hybrid Spatio-Spectral Regularization |
| title_full_unstemmed | A Constrained Convex Optimization Approach to Hyperspectral Image Restoration with Hybrid Spatio-Spectral Regularization |
| title_short | A Constrained Convex Optimization Approach to Hyperspectral Image Restoration with Hybrid Spatio-Spectral Regularization |
| title_sort | constrained convex optimization approach to hyperspectral image restoration with hybrid spatio spectral regularization |
| topic | hyperspectral image restoration ADMM mixed noise removal compressed sensing |
| url | https://www.mdpi.com/2072-4292/12/21/3541 |
| work_keys_str_mv | AT saoritakeyama aconstrainedconvexoptimizationapproachtohyperspectralimagerestorationwithhybridspatiospectralregularization AT shunsukeono aconstrainedconvexoptimizationapproachtohyperspectralimagerestorationwithhybridspatiospectralregularization AT itsuokumazawa aconstrainedconvexoptimizationapproachtohyperspectralimagerestorationwithhybridspatiospectralregularization AT saoritakeyama constrainedconvexoptimizationapproachtohyperspectralimagerestorationwithhybridspatiospectralregularization AT shunsukeono constrainedconvexoptimizationapproachtohyperspectralimagerestorationwithhybridspatiospectralregularization AT itsuokumazawa constrainedconvexoptimizationapproachtohyperspectralimagerestorationwithhybridspatiospectralregularization |