Deep Residual Haze Network for Image Dehazing and Deraining
Image dehazing on a hazy image aims to remove the haze and make the image scene clear, which attracts more and more research interests in recent years. Most existing image dehazing methods use a classic atmospheric scattering model and natural image priors to remove the image haze. In this paper, we...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2020-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8957649/ |
id |
doaj-fbb32556ad194cd882d5951260ed51b0 |
---|---|
record_format |
Article |
spelling |
doaj-fbb32556ad194cd882d5951260ed51b02021-03-30T01:49:14ZengIEEEIEEE Access2169-35362020-01-0189488950010.1109/ACCESS.2020.29642718957649Deep Residual Haze Network for Image Dehazing and DerainingChuansheng Wang0https://orcid.org/0000-0003-4340-7831Zuoyong Li1https://orcid.org/0000-0003-0952-9915Jiawei Wu2https://orcid.org/0000-0001-6251-2202Haoyi Fan3https://orcid.org/0000-0001-9428-7812Guobao Xiao4https://orcid.org/0000-0003-2928-8100Hong Zhang5https://orcid.org/0000-0002-1347-9116Fujian Provincial Key Laboratory of Information Processing and Intelligent Control, Minjiang University, Fuzhou, ChinaFujian Provincial Key Laboratory of Information Processing and Intelligent Control, Minjiang University, Fuzhou, ChinaFujian Provincial Key Laboratory of Information Processing and Intelligent Control, Minjiang University, Fuzhou, ChinaSchool of Computer Science and Technology, Harbin University of Science and Technology, Harbin, ChinaFujian Provincial Key Laboratory of Information Processing and Intelligent Control, Minjiang University, Fuzhou, ChinaSchool of Computer Science and Technology, Harbin University of Science and Technology, Harbin, ChinaImage dehazing on a hazy image aims to remove the haze and make the image scene clear, which attracts more and more research interests in recent years. Most existing image dehazing methods use a classic atmospheric scattering model and natural image priors to remove the image haze. In this paper, we propose an end-to-end image dehazing model termed as DRHNet (Deep Residual Haze Network), which restores the haze-free image by subtracting the learned negative residual map from the hazy image. Specifically, DRHNet proposes a context-aware feature extraction module to aggregate the contextual information effectively. Furthermore, it proposes a novel nonlinear activation function termed as RPReLU (Reverse Parametric Rectified Linear Unit) to improve its representation ability and to accelerate its convergence. Extensive experiments demonstrate that DRHNet outperforms state-of-the-art methods both quantitatively and qualitatively. In addition, experiments on image deraining task show that DRHNet can also serve for image deraining.https://ieeexplore.ieee.org/document/8957649/Image dehazingimage derainingnegative residual mapcontext-aware feature extractionreverse parametric rectified linear unit (RPReLU) |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Chuansheng Wang Zuoyong Li Jiawei Wu Haoyi Fan Guobao Xiao Hong Zhang |
spellingShingle |
Chuansheng Wang Zuoyong Li Jiawei Wu Haoyi Fan Guobao Xiao Hong Zhang Deep Residual Haze Network for Image Dehazing and Deraining IEEE Access Image dehazing image deraining negative residual map context-aware feature extraction reverse parametric rectified linear unit (RPReLU) |
author_facet |
Chuansheng Wang Zuoyong Li Jiawei Wu Haoyi Fan Guobao Xiao Hong Zhang |
author_sort |
Chuansheng Wang |
title |
Deep Residual Haze Network for Image Dehazing and Deraining |
title_short |
Deep Residual Haze Network for Image Dehazing and Deraining |
title_full |
Deep Residual Haze Network for Image Dehazing and Deraining |
title_fullStr |
Deep Residual Haze Network for Image Dehazing and Deraining |
title_full_unstemmed |
Deep Residual Haze Network for Image Dehazing and Deraining |
title_sort |
deep residual haze network for image dehazing and deraining |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2020-01-01 |
description |
Image dehazing on a hazy image aims to remove the haze and make the image scene clear, which attracts more and more research interests in recent years. Most existing image dehazing methods use a classic atmospheric scattering model and natural image priors to remove the image haze. In this paper, we propose an end-to-end image dehazing model termed as DRHNet (Deep Residual Haze Network), which restores the haze-free image by subtracting the learned negative residual map from the hazy image. Specifically, DRHNet proposes a context-aware feature extraction module to aggregate the contextual information effectively. Furthermore, it proposes a novel nonlinear activation function termed as RPReLU (Reverse Parametric Rectified Linear Unit) to improve its representation ability and to accelerate its convergence. Extensive experiments demonstrate that DRHNet outperforms state-of-the-art methods both quantitatively and qualitatively. In addition, experiments on image deraining task show that DRHNet can also serve for image deraining. |
topic |
Image dehazing image deraining negative residual map context-aware feature extraction reverse parametric rectified linear unit (RPReLU) |
url |
https://ieeexplore.ieee.org/document/8957649/ |
work_keys_str_mv |
AT chuanshengwang deepresidualhazenetworkforimagedehazingandderaining AT zuoyongli deepresidualhazenetworkforimagedehazingandderaining AT jiaweiwu deepresidualhazenetworkforimagedehazingandderaining AT haoyifan deepresidualhazenetworkforimagedehazingandderaining AT guobaoxiao deepresidualhazenetworkforimagedehazingandderaining AT hongzhang deepresidualhazenetworkforimagedehazingandderaining |
_version_ |
1724186392675745792 |