Image Denoising via Multi-Scale Gated Fusion Network
Deep convolutional neural networks have made significant progress in image denoising. However, in most cases, denoising methods using a single-stream structure with a single kernel size do not perform so well in integrating complementary contextual information; owing to the lack of this type of info...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2019-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8689013/ |
id |
doaj-319751b48bfc44268db85f3609a8dca3 |
---|---|
record_format |
Article |
spelling |
doaj-319751b48bfc44268db85f3609a8dca32021-03-29T22:18:07ZengIEEEIEEE Access2169-35362019-01-017493924940210.1109/ACCESS.2019.29108798689013Image Denoising via Multi-Scale Gated Fusion NetworkShengyu Li0https://orcid.org/0000-0002-4907-4014Yaowu Chen1Rongxin Jiang2Xiang Tian3College of Biomedical Engineering & Instrument Science, Zhejiang University, Hangzhou, ChinaZhejiang University Embedded System Engineering Research Center, Ministry of Education of China, Zhejiang University, Hangzhou, ChinaThe State Key Laboratory of Industrial Control Technology, Zhejiang University, Hangzhou, ChinaZhejiang Provincial Key Laboratory for Network Multimedia Technologies, Zhejiang University, Hangzhou, ChinaDeep convolutional neural networks have made significant progress in image denoising. However, in most cases, denoising methods using a single-stream structure with a single kernel size do not perform so well in integrating complementary contextual information; owing to the lack of this type of information, they may fail to reconstruct fine textures and patterns. To address this problem, we propose a multi-scale gated fusion network (MGFN) for image denoising, which learns direct end-to-end mappings from corrupted images to clean images. Our proposed network consists of several multi-scale mutually-gated (MM) blocks. In each MM block, we incorporate dilated convolution into a merge-and-run (MR) module to exploit multi-scale features in an effective way and further recognize useful features by filtration via a gating mechanism. Moreover, we propose a simple but effective loss function named dropout-loss to train the network. The extensive experiments on benchmark datasets show that our proposed method can well recover textures, yielding favorable performance against other state-of-the-art methods.https://ieeexplore.ieee.org/document/8689013/Image denoisingconvolutional neural networksmerge-and-runmulti-scale |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Shengyu Li Yaowu Chen Rongxin Jiang Xiang Tian |
spellingShingle |
Shengyu Li Yaowu Chen Rongxin Jiang Xiang Tian Image Denoising via Multi-Scale Gated Fusion Network IEEE Access Image denoising convolutional neural networks merge-and-run multi-scale |
author_facet |
Shengyu Li Yaowu Chen Rongxin Jiang Xiang Tian |
author_sort |
Shengyu Li |
title |
Image Denoising via Multi-Scale Gated Fusion Network |
title_short |
Image Denoising via Multi-Scale Gated Fusion Network |
title_full |
Image Denoising via Multi-Scale Gated Fusion Network |
title_fullStr |
Image Denoising via Multi-Scale Gated Fusion Network |
title_full_unstemmed |
Image Denoising via Multi-Scale Gated Fusion Network |
title_sort |
image denoising via multi-scale gated fusion network |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2019-01-01 |
description |
Deep convolutional neural networks have made significant progress in image denoising. However, in most cases, denoising methods using a single-stream structure with a single kernel size do not perform so well in integrating complementary contextual information; owing to the lack of this type of information, they may fail to reconstruct fine textures and patterns. To address this problem, we propose a multi-scale gated fusion network (MGFN) for image denoising, which learns direct end-to-end mappings from corrupted images to clean images. Our proposed network consists of several multi-scale mutually-gated (MM) blocks. In each MM block, we incorporate dilated convolution into a merge-and-run (MR) module to exploit multi-scale features in an effective way and further recognize useful features by filtration via a gating mechanism. Moreover, we propose a simple but effective loss function named dropout-loss to train the network. The extensive experiments on benchmark datasets show that our proposed method can well recover textures, yielding favorable performance against other state-of-the-art methods. |
topic |
Image denoising convolutional neural networks merge-and-run multi-scale |
url |
https://ieeexplore.ieee.org/document/8689013/ |
work_keys_str_mv |
AT shengyuli imagedenoisingviamultiscalegatedfusionnetwork AT yaowuchen imagedenoisingviamultiscalegatedfusionnetwork AT rongxinjiang imagedenoisingviamultiscalegatedfusionnetwork AT xiangtian imagedenoisingviamultiscalegatedfusionnetwork |
_version_ |
1724191825937301504 |