Single image haze removal algorithm using pixel-based airlight constraints

Zhenwei Gao*, Yongqiang Bai

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

9 Citations (Scopus)

Abstract

Fast single image dehazing has been a challenging problem in many fields, such as computer vision and real-time applications. The existing image dehazing algorithms cannot achieve a trade-off between the dehazing performance and the computational complexity. The proposed approach first applies the mean filter twice to estimate airlight, which include pixel-based dark channel and bright channel constraints. And then the relationship between channel values of the restored image and atmospheric light is qualitatively analyzed to give the optimum estimate of atmospheric light. Using the airlight and atmospheric light, we can easily restore the scene radiance via the atmospheric scattering model. Compared with others, the main advantage of the proposed approach is its high speed and significant visibility improvement even in the sky and white areas. This speed allows the enhanced haze image to be applied in real-time processing applications. A comparative study and quantitative evaluation are proposed with a few other state of the art algorithms which demonstrates that similar or better quality results are obtained.

Original languageEnglish
Title of host publication2016 22nd International Conference on Automation and Computing, ICAC 2016
Subtitle of host publicationTackling the New Challenges in Automation and Computing
EditorsJing Wang, Zhijie Xu
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages267-272
Number of pages6
ISBN (Electronic)9781862181311
DOIs
Publication statusPublished - 20 Oct 2016
Event22nd International Conference on Automation and Computing, ICAC 2016 - Colchester, United Kingdom
Duration: 7 Sept 20168 Sept 2016

Publication series

Name2016 22nd International Conference on Automation and Computing, ICAC 2016: Tackling the New Challenges in Automation and Computing

Conference

Conference22nd International Conference on Automation and Computing, ICAC 2016
Country/TerritoryUnited Kingdom
CityColchester
Period7/09/168/09/16

Keywords

  • airlight constraints
  • dehazing
  • mean filter
  • real-time

Fingerprint

Dive into the research topics of 'Single image haze removal algorithm using pixel-based airlight constraints'. Together they form a unique fingerprint.

Cite this