文献类型：期刊文献

英文题名：Simplified non-locally dense network for single-image dehazing

作者：Chen, Zhihua[1];Hu, Zhuoliang[1];Sheng, Bin[2];Li, Ping[3];Kim, Jinman[4];Wu, Enhua[5,6]

机构：[1]East China Univ Sci & Technol, Dept Comp Sci & Engn, Shanghai, Peoples R China;[2]Shanghai Jiao Tong Univ, Dept Comp Sci & Engn, Shanghai, Peoples R China;[3]Hong Kong Polytech Univ, Dept Comp, Kowloon, Hong Kong, Peoples R China;[4]Univ Sydney, Sch Informat Technol, Biomed & Multimedia Informat Technol Res Grp, Sydney, NSW, Australia;[5]Chinese Acad Sci, Inst Software, State Key Lab Comp Sci, Beijing, Peoples R China;[6]Univ Macau, Fac Sci & Technol, Macau, Peoples R China

年份：2020

卷号：36

期号：10-12

起止页码：2189

外文期刊名：VISUAL COMPUTER

收录：;EI(收录号：20203209012012);WOS:【SCI-EXPANDED(收录号:WOS:000551368400004)】；

基金：This work was supported in part by the National Natural Science Foundation of China under Grant 61672228, Grant 61872241, Grant 61572316, and Grant 61370174, in part by the Shanghai Automotive Industry Science and Technology Development Foundation under Grant 1837, and in part by The Hong Kong Polytechnic University under Grant P0030419 and Grant P0030929.

语种：英文

外文关键词：Single-image dehazing; Dense; Non-local

摘要：Single-image dehazing is an ill-posed problem. Most previous methods focused on estimating intermediate parameters for input hazy images. In this paper, we propose a novel end-to-end Simplified Non-locally Dense Network (SNDN) which does not rely on intermediate parameters. To capture long-range dependencies, we propose a Simplified Non-local Dense Block (SNDB) which is lightweight and outperforms traditional non-local method. Our SNDB will be embedded into a densely connected encoder-decoder network. To avoid gradients vanishing problem, we propose a simple branch network which only have five convolution layers. The effectiveness of our proposed network is proved through ablation experiment. In addition, we enhanced our training set by synthesizing colored hazy images, which helps restore the original color of the hazy image. The experimental results demonstrate that our network have better performance than most of the pervious state-of-the-art methods.