Residual Dense Blocks and Contrastive Regularization Integrated Underwater Image Enhancement Network
Due to severe information degradation, underwater image deblurring remains a challenging ill-posed problem. However, most deep learning models do not adequately exploit the hierarchical features of the original underwater images. They typically use clear images as positive samples to guide the train...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10274961/ |
| _version_ | 1797655643713175552 |
|---|---|
| author | Hualong Zhao Hongchun Yuan |
| author_facet | Hualong Zhao Hongchun Yuan |
| author_sort | Hualong Zhao |
| collection | DOAJ |
| description | Due to severe information degradation, underwater image deblurring remains a challenging ill-posed problem. However, most deep learning models do not adequately exploit the hierarchical features of the original underwater images. They typically use clear images as positive samples to guide the training of image enhancement networks, while neglecting the utilization of negative information. In this paper, we introduce the Residual Dense Block (RDB) and Contrastive Regularization (CR) techniques. By leveraging the local and global feature fusion of RDB and the contrastive learning of CR, our model effectively extracts multi-level features from the original images, adaptively preserves hierarchical features, and achieves high-quality underwater image deblurring through learning from the original images. Experimental results demonstrate that our model outperforms other comparative algorithms in terms of subjective visual quality and objective evaluation metrics across four datasets. |
| first_indexed | 2024-03-11T17:17:27Z |
| format | Article |
| id | doaj.art-63b5496c33e348fcbaae3e4410a7c306 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-11T17:17:27Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-63b5496c33e348fcbaae3e4410a7c3062023-10-19T23:01:33ZengIEEEIEEE Access2169-35362023-01-011111301711302610.1109/ACCESS.2023.332336010274961Residual Dense Blocks and Contrastive Regularization Integrated Underwater Image Enhancement NetworkHualong Zhao0https://orcid.org/0009-0009-9502-5510Hongchun Yuan1https://orcid.org/0000-0002-6869-9159College of Information Technology, Shanghai Ocean University, Shanghai, ChinaCollege of Information Technology, Shanghai Ocean University, Shanghai, ChinaDue to severe information degradation, underwater image deblurring remains a challenging ill-posed problem. However, most deep learning models do not adequately exploit the hierarchical features of the original underwater images. They typically use clear images as positive samples to guide the training of image enhancement networks, while neglecting the utilization of negative information. In this paper, we introduce the Residual Dense Block (RDB) and Contrastive Regularization (CR) techniques. By leveraging the local and global feature fusion of RDB and the contrastive learning of CR, our model effectively extracts multi-level features from the original images, adaptively preserves hierarchical features, and achieves high-quality underwater image deblurring through learning from the original images. Experimental results demonstrate that our model outperforms other comparative algorithms in terms of subjective visual quality and objective evaluation metrics across four datasets.https://ieeexplore.ieee.org/document/10274961/Underwater image enhancementdeep learning methodsresidual dense network |
| spellingShingle | Hualong Zhao Hongchun Yuan Residual Dense Blocks and Contrastive Regularization Integrated Underwater Image Enhancement Network IEEE Access Underwater image enhancement deep learning methods residual dense network |
| title | Residual Dense Blocks and Contrastive Regularization Integrated Underwater Image Enhancement Network |
| title_full | Residual Dense Blocks and Contrastive Regularization Integrated Underwater Image Enhancement Network |
| title_fullStr | Residual Dense Blocks and Contrastive Regularization Integrated Underwater Image Enhancement Network |
| title_full_unstemmed | Residual Dense Blocks and Contrastive Regularization Integrated Underwater Image Enhancement Network |
| title_short | Residual Dense Blocks and Contrastive Regularization Integrated Underwater Image Enhancement Network |
| title_sort | residual dense blocks and contrastive regularization integrated underwater image enhancement network |
| topic | Underwater image enhancement deep learning methods residual dense network |
| url | https://ieeexplore.ieee.org/document/10274961/ |
| work_keys_str_mv | AT hualongzhao residualdenseblocksandcontrastiveregularizationintegratedunderwaterimageenhancementnetwork AT hongchunyuan residualdenseblocksandcontrastiveregularizationintegratedunderwaterimageenhancementnetwork |