Dehazing Based on Long-Range Dependence of Foggy Images

Deep neural networks (DNNs) with long-range dependence (LRD) have attracted more and more attention recently. However, LRD of DNNs is proposed from the view on gradient disappearance in training, which lacks theory analysis. In order to prove LRD of foggy images, the Hurst parameters of over 1,000 f...

Full description

Saved in:
Bibliographic Details
Published inFrontiers in physics Vol. 10
Main Authors Yuan, Hong Xu, Liao, Zhiwu, Wang, Rui Xin, Dong, Xinceng, Liu, Tao, Long, Wu Dan, Wei, Qing Jin, Xu, Ya Jie, Yu, Yong, Chen, Peng, Hou, Rong
Format Journal Article
LanguageEnglish
Published Frontiers Media S.A 16.02.2022
Subjects
deep neural network
Hurst parameter (H)
image dehazing
long-range dependence
residual dense block
residual dense block group
Online AccessGet full text

Cover

More Information
Summary:Deep neural networks (DNNs) with long-range dependence (LRD) have attracted more and more attention recently. However, LRD of DNNs is proposed from the view on gradient disappearance in training, which lacks theory analysis. In order to prove LRD of foggy images, the Hurst parameters of over 1,000 foggy images in SOTS are computed and discussed. Then, the Residual Dense Block Group (RDBG), which has additional long skips among two Residual Dense Blocks to fit LRD of foggy images, is proposed. The Residual Dense Block Group can significantly improve the details of dehazing image in dense fog and reduce the artifacts of dehazing image.
ISSN:2296-424X
2296-424X
DOI:10.3389/fphy.2022.828804