Low Light Image Enhancement for Color Images Combined with Sky Region Segmentation

  • Chao Fang
  • , Changfeng Lv
  • , Fudong Cai
  • , Huanyun Liu
  • , Jinjun Wang
  • , Minwei Shuai

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

3 Scopus citations

Abstract

To solve the problems of low contrast and blurred details in low light images, a brand-new low light image enhancement algorithm combined with sky region segmentation is proposed for color images in this paper. In the proposed algorithm, the image enhancement is processed in HSV space to achieve color consistency. An adaptive global enhancement method is adopted in brightness enhancement. In addition, a simple nonlinear transformation is applied on the saturation component combined with sky region segmentation. Experimental results verify the effectiveness of the proposed algorithm, and reveal the proposed algorithm could improve the brightness, the contrast and details of low light images, while maintaining the color consistency to make the images more in line with the visual effects of human eyes.

Original languageEnglish
Title of host publicationProceedings - 2022 International Conference on Machine Learning and Knowledge Engineering, MLKE 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages169-172
Number of pages4
ISBN (Electronic)9781665495677
DOIs
StatePublished - 2022
Event2022 International Conference on Machine Learning and Knowledge Engineering, MLKE 2022 - Virtual, Guilin, China
Duration: 25 Feb 202227 Feb 2022

Publication series

NameProceedings - 2022 International Conference on Machine Learning and Knowledge Engineering, MLKE 2022

Conference

Conference2022 International Conference on Machine Learning and Knowledge Engineering, MLKE 2022
Country/TerritoryChina
CityVirtual, Guilin
Period25/02/2227/02/22

Keywords

  • Sky segmentation
  • image enhancement
  • low light image

Fingerprint

Dive into the research topics of 'Low Light Image Enhancement for Color Images Combined with Sky Region Segmentation'. Together they form a unique fingerprint.

Cite this