Single Image Dehazing via Deep Learning-based Image Restoration

Chia Hung Yeh, Chih Hsiang Huang, Li Wei Kang, Min Hui Lin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Images/videos captured from outdoor visual devices are usually degraded by turbid media, such as haze, smoke, fog, rain, and snow. Haze is the most common one in outdoor scenes due to the atmosphere conditions. This paper presents a deep learning-based architecture for single image dehazing via image restoration. instead of learning an end-to-end mapping between each pair of hazy image and its corresponding haze-free one adopted by most existing approaches, we propose to transform the problem into the restoration of the image base component. By first decomposing the hazy image into the base and the detail components, haze removal can be achieved by learning a CNN (convolutional neural network) only for mapping between hazy and haze-free base components, while the detail component can be further enhanced. As a result, the final dehazed image is obtained by integrating the haze-removed base and the enhanced detail image components. Experimental results have demonstrated good efficacy of the proposed method, compared with state-of-the-art results.

Original languageEnglish
Title of host publication2018 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2018 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1609-1615
Number of pages7
ISBN (Electronic)9789881476852
DOIs
Publication statusPublished - 2019 Mar 4
Event10th Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2018 - Honolulu, United States
Duration: 2018 Nov 122018 Nov 15

Publication series

Name2018 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2018 - Proceedings

Conference

Conference10th Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC 2018
Country/TerritoryUnited States
CityHonolulu
Period2018/11/122018/11/15

ASJC Scopus subject areas

  • Information Systems

Fingerprint

Dive into the research topics of 'Single Image Dehazing via Deep Learning-based Image Restoration'. Together they form a unique fingerprint.

Cite this