SIA OpenIR  > 机器人学研究室
Deep Retinex Network for Single Image Dehazing
Li PY(李鹏越)1,2,3; Tian JD(田建东)2,3,4; Tang YD(唐延东)2,3,4; Wang GL(王国霖)5; Wu CD(吴成东)1
Source PublicationIEEE Transactions on Image Processing
Indexed BySCI ; EI
EI Accession number20205009620129
WOS IDWOS:000600285000001
Contribution Rank1
Funding OrganizationNatural Science Foundation of China under Grant U2013210 and Grant 61821005 ; LiaoNing Revitalization Talents Program under Grant XLYC 1907039 ; Youth Innovation Promotion Association CAS
KeywordImage dehazing retinex theory pixel-wise attention image restoration

In this paper, we propose a retinex-based decomposition model for a hazy image and a novel end-to-end image dehazing network. In the model, the illumination of the hazy image is decomposed into natural illumination for the haze-free image and residual illumination caused by haze. Based on this model, we design a deep retinex dehazing network (RDN) to jointly estimate the residual illumination map and the haze-free image. Our RDN consists of a multiscale residual dense network for estimating the residual illumination map and a U-Net with channel and spatial attention mechanisms for image dehazing. The multiscale residual dense network can simultaneously capture global contextual information from small-scale receptive fields and local detailed information from large-scale receptive fields to precisely estimate the residual illumination map caused by haze. In the dehazing U-Net, we apply the channel and spatial attention mechanisms in the skip connection of the U-Net to achieve a trade-off between overdehazing and underdehazing by automatically adjusting the channel-wise and pixel-wise attention weights. Compared with scattering model-based networks, fully data-driven networks, and prior-based dehazing methods, our RDN can avoid the errors associated with the simplified scattering model and provide better generalization ability with no dependence on prior information. Extensive experiments show the superiority of the RDN to various state-of-the-art methods.

WOS SubjectComputer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS Research AreaComputer Science ; Engineering
Funding ProjectNatural Science Foundation of China[U2013210] ; Natural Science Foundation of China[61821005] ; LiaoNing Revitalization Talents Program[XLYC 1907039] ; Youth Innovation Promotion Association CAS
Citation statistics
Cited Times:1[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Corresponding AuthorTian JD(田建东)
Affiliation1.Faculty of Robot Science and Engineering, Northeastern University, Shenyang, China, 110004
2.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China, 110016
3.Institutes for Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, China,110169
4.College of Robotics and Intelligent Manufacturing, the University of Chinese Academy of Sciences, Beijing, China, 100049
5.College of Computer Science and Technology, Jilin University, Changchun, China, 130012
Recommended Citation
GB/T 7714
Li PY,Tian JD,Tang YD,et al. Deep Retinex Network for Single Image Dehazing[J]. IEEE Transactions on Image Processing,2021,30:1100-1115.
APA Li PY,Tian JD,Tang YD,Wang GL,&Wu CD.(2021).Deep Retinex Network for Single Image Dehazing.IEEE Transactions on Image Processing,30,1100-1115.
MLA Li PY,et al."Deep Retinex Network for Single Image Dehazing".IEEE Transactions on Image Processing 30(2021):1100-1115.
Files in This Item:
File Name/Size DocType Version Access License
Deep Retinex Network(8798KB)期刊论文出版稿开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Li PY(李鹏越)]'s Articles
[Tian JD(田建东)]'s Articles
[Tang YD(唐延东)]'s Articles
Baidu academic
Similar articles in Baidu academic
[Li PY(李鹏越)]'s Articles
[Tian JD(田建东)]'s Articles
[Tang YD(唐延东)]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Li PY(李鹏越)]'s Articles
[Tian JD(田建东)]'s Articles
[Tang YD(唐延东)]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Deep Retinex Network for Single Image Dehazing.pdf
Format: Adobe PDF
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.