Ensemble Dictionary Learning for Single Image Deblurring via Low-Rank Regularization
Sparse representation is a powerful statistical technique that has been widely utilized in image restoration applications. In this paper, an improved sparse representation model regularized by a low-rank constraint is proposed for single image deblurring. The key motivation for the proposed model li...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-03-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1424-8220/19/5/1143 |
| _version_ | 1811184832456163328 |
|---|---|
| author | Jinyang Li Zhijing Liu |
| author_facet | Jinyang Li Zhijing Liu |
| author_sort | Jinyang Li |
| collection | DOAJ |
| description | Sparse representation is a powerful statistical technique that has been widely utilized in image restoration applications. In this paper, an improved sparse representation model regularized by a low-rank constraint is proposed for single image deblurring. The key motivation for the proposed model lies in the observation that natural images are full of self-repetitive structures and they can be represented by similar patterns. However, as input images contain noise, blur, and other visual artifacts, extracting nonlocal similarities only with patch clustering algorithms is insufficient. In this paper, we first propose an ensemble dictionary learning method to represent different similar patterns. Then, low-rank embedded regularization is directly imposed on inputs to regularize the desired solution space which favors natural and sharp structures. The proposed method can be optimized by alternatively solving nuclear norm minimization and l 1 norm minimization problems to achieve higher restoration quality. Experimental comparisons validate the superior results of the proposed method compared with other deblurring algorithms in terms of visual quality and quantitative metrics. |
| first_indexed | 2024-04-11T13:19:12Z |
| format | Article |
| id | doaj.art-66bf0032696d460cb5cc541101e7e091 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-11T13:19:12Z |
| publishDate | 2019-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-66bf0032696d460cb5cc541101e7e0912022-12-22T04:22:16ZengMDPI AGSensors1424-82202019-03-01195114310.3390/s19051143s19051143Ensemble Dictionary Learning for Single Image Deblurring via Low-Rank RegularizationJinyang Li0Zhijing Liu1School of Computer Science and Technology, Xidian University, Xi’an 710071, ChinaSchool of Computer Science and Technology, Xidian University, Xi’an 710071, ChinaSparse representation is a powerful statistical technique that has been widely utilized in image restoration applications. In this paper, an improved sparse representation model regularized by a low-rank constraint is proposed for single image deblurring. The key motivation for the proposed model lies in the observation that natural images are full of self-repetitive structures and they can be represented by similar patterns. However, as input images contain noise, blur, and other visual artifacts, extracting nonlocal similarities only with patch clustering algorithms is insufficient. In this paper, we first propose an ensemble dictionary learning method to represent different similar patterns. Then, low-rank embedded regularization is directly imposed on inputs to regularize the desired solution space which favors natural and sharp structures. The proposed method can be optimized by alternatively solving nuclear norm minimization and l 1 norm minimization problems to achieve higher restoration quality. Experimental comparisons validate the superior results of the proposed method compared with other deblurring algorithms in terms of visual quality and quantitative metrics.http://www.mdpi.com/1424-8220/19/5/1143image deblurringlow-rank constraintself-repetitive structuresnonlocal similarityensemble dictionary learning |
| spellingShingle | Jinyang Li Zhijing Liu Ensemble Dictionary Learning for Single Image Deblurring via Low-Rank Regularization Sensors image deblurring low-rank constraint self-repetitive structures nonlocal similarity ensemble dictionary learning |
| title | Ensemble Dictionary Learning for Single Image Deblurring via Low-Rank Regularization |
| title_full | Ensemble Dictionary Learning for Single Image Deblurring via Low-Rank Regularization |
| title_fullStr | Ensemble Dictionary Learning for Single Image Deblurring via Low-Rank Regularization |
| title_full_unstemmed | Ensemble Dictionary Learning for Single Image Deblurring via Low-Rank Regularization |
| title_short | Ensemble Dictionary Learning for Single Image Deblurring via Low-Rank Regularization |
| title_sort | ensemble dictionary learning for single image deblurring via low rank regularization |
| topic | image deblurring low-rank constraint self-repetitive structures nonlocal similarity ensemble dictionary learning |
| url | http://www.mdpi.com/1424-8220/19/5/1143 |
| work_keys_str_mv | AT jinyangli ensembledictionarylearningforsingleimagedeblurringvialowrankregularization AT zhijingliu ensembledictionarylearningforsingleimagedeblurringvialowrankregularization |