PDR-Net : Perception-Inspired Single Image Dehazing Network with Refinement

Chongyi LI, Chunle GUO*, Jichang GUO, Ping HAN, Huazhu FU, Runmin CONG

*Corresponding author for this work

Research output: Journal PublicationsJournal Article (refereed)peer-review

113 Citations (Scopus)

Abstract

During recent years, we have witnessed a rapid development of wireless network technologies which have revolutionized the way people take and share multimedia content. However, images captured in the outdoor scenes usually suffer from limited visibility due to suspended atmospheric particles, which directly affects the quality of photos. Despite the recent progress of image dehazing methods, the visual quality of dehazed results still needs further improvement. In this paper, we propose a deep convolutional neural network (CNN) for single image dehazing called PDR-Net, which includes a perception-inspired haze removal subnetwork that reconstructs the latent dehazed image and a refinement subnetwork that further enhances the contrast and color properties of the dehazed result by joint multi-term loss optimization. Compared to the previous methods, our method combines the advantages of existing indoor and outdoor image dehazing training data, which makes the proposed PDR-Net generalized to various hazy images and effective for improving the visual quality of the dehazed results. Extensive experiments demonstrate that the proposed method achieves comparable and even better performance on both real and synthetic images in qualitative and quantitative metrics. Additionally, the potential usage of our method in high-level vision tasks is discussed.

Original languageEnglish
Article number8792133
Pages (from-to)704-716
Number of pages13
JournalIEEE Transactions on Multimedia
Volume22
Issue number3
Early online date8 Aug 2019
DOIs
Publication statusPublished - Mar 2020
Externally publishedYes

Bibliographical note

Publisher Copyright:
© 1999-2012 IEEE.

Keywords

  • end-to-end
  • Image dehazing
  • image refinement
  • perceptual loss

Fingerprint

Dive into the research topics of 'PDR-Net : Perception-Inspired Single Image Dehazing Network with Refinement'. Together they form a unique fingerprint.

Cite this