Iterative deep neural networks based on proximal gradient descent for image restoration.
The algorithm unfolding networks with explainability of algorithms and higher efficiency of Deep Neural Networks (DNN) have received considerable attention in solving ill-posed inverse problems. Under the algorithm unfolding network framework, we propose a novel end-to-end iterative deep neural netw...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2022-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0276373 |
| _version_ | 1811231007424118784 |
|---|---|
| author | Ting Lv Zhenkuan Pan Weibo Wei Guangyu Yang Jintao Song Xuqing Wang Lu Sun Qian Li Xiatao Sun |
| author_facet | Ting Lv Zhenkuan Pan Weibo Wei Guangyu Yang Jintao Song Xuqing Wang Lu Sun Qian Li Xiatao Sun |
| author_sort | Ting Lv |
| collection | DOAJ |
| description | The algorithm unfolding networks with explainability of algorithms and higher efficiency of Deep Neural Networks (DNN) have received considerable attention in solving ill-posed inverse problems. Under the algorithm unfolding network framework, we propose a novel end-to-end iterative deep neural network and its fast network for image restoration. The first one is designed making use of proximal gradient descent algorithm of variational models, which consists of denoiser and reconstruction sub-networks. The second one is its accelerated version with momentum factors. For sub-network of denoiser, we embed the Convolutional Block Attention Module (CBAM) in previous U-Net for adaptive feature refinement. Experiments on image denoising and deblurring demonstrate that competitive performances in quality and efficiency are gained by compared with several state-of-the-art networks for image restoration. Proposed unfolding DNN can be easily extended to solve other similar image restoration tasks, such as image super-resolution, image demosaicking, etc. |
| first_indexed | 2024-04-12T10:38:58Z |
| format | Article |
| id | doaj.art-241bebf3cbf649ecacfe37c25785e443 |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-04-12T10:38:58Z |
| publishDate | 2022-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-241bebf3cbf649ecacfe37c25785e4432022-12-22T03:36:39ZengPublic Library of Science (PLoS)PLoS ONE1932-62032022-01-011711e027637310.1371/journal.pone.0276373Iterative deep neural networks based on proximal gradient descent for image restoration.Ting LvZhenkuan PanWeibo WeiGuangyu YangJintao SongXuqing WangLu SunQian LiXiatao SunThe algorithm unfolding networks with explainability of algorithms and higher efficiency of Deep Neural Networks (DNN) have received considerable attention in solving ill-posed inverse problems. Under the algorithm unfolding network framework, we propose a novel end-to-end iterative deep neural network and its fast network for image restoration. The first one is designed making use of proximal gradient descent algorithm of variational models, which consists of denoiser and reconstruction sub-networks. The second one is its accelerated version with momentum factors. For sub-network of denoiser, we embed the Convolutional Block Attention Module (CBAM) in previous U-Net for adaptive feature refinement. Experiments on image denoising and deblurring demonstrate that competitive performances in quality and efficiency are gained by compared with several state-of-the-art networks for image restoration. Proposed unfolding DNN can be easily extended to solve other similar image restoration tasks, such as image super-resolution, image demosaicking, etc.https://doi.org/10.1371/journal.pone.0276373 |
| spellingShingle | Ting Lv Zhenkuan Pan Weibo Wei Guangyu Yang Jintao Song Xuqing Wang Lu Sun Qian Li Xiatao Sun Iterative deep neural networks based on proximal gradient descent for image restoration. PLoS ONE |
| title | Iterative deep neural networks based on proximal gradient descent for image restoration. |
| title_full | Iterative deep neural networks based on proximal gradient descent for image restoration. |
| title_fullStr | Iterative deep neural networks based on proximal gradient descent for image restoration. |
| title_full_unstemmed | Iterative deep neural networks based on proximal gradient descent for image restoration. |
| title_short | Iterative deep neural networks based on proximal gradient descent for image restoration. |
| title_sort | iterative deep neural networks based on proximal gradient descent for image restoration |
| url | https://doi.org/10.1371/journal.pone.0276373 |
| work_keys_str_mv | AT tinglv iterativedeepneuralnetworksbasedonproximalgradientdescentforimagerestoration AT zhenkuanpan iterativedeepneuralnetworksbasedonproximalgradientdescentforimagerestoration AT weibowei iterativedeepneuralnetworksbasedonproximalgradientdescentforimagerestoration AT guangyuyang iterativedeepneuralnetworksbasedonproximalgradientdescentforimagerestoration AT jintaosong iterativedeepneuralnetworksbasedonproximalgradientdescentforimagerestoration AT xuqingwang iterativedeepneuralnetworksbasedonproximalgradientdescentforimagerestoration AT lusun iterativedeepneuralnetworksbasedonproximalgradientdescentforimagerestoration AT qianli iterativedeepneuralnetworksbasedonproximalgradientdescentforimagerestoration AT xiataosun iterativedeepneuralnetworksbasedonproximalgradientdescentforimagerestoration |