Real image dehazing based on model optimization and luminance enhancement

Addressing the issues of blurred details, color distortion, and reduced brightness in hazy images, we propose a dehazing method grounded in model optimization and brightness enhancement. Firstly, the traditional atmospheric scattering model is optimized based on the dark channel prior theory. Second...

Full description

Saved in:
Bibliographic Details
Published inScientific Bulletin. Series C, Electrical Engineering and Computer Science no. 1; p. 167
Main Authors Li, Lingyu, Tao, Zhiyong, Lin, Sen
Format Journal Article
LanguageEnglish
Published Bucharest University Polytechnica of Bucharest 01.01.2025
Subjects
Atmospheric scattering
Brightness
Color
Image quality
Luminance
Optimization
Online AccessGet full text
ISSN2286-3540

Cover

More Information
Summary:Addressing the issues of blurred details, color distortion, and reduced brightness in hazy images, we propose a dehazing method grounded in model optimization and brightness enhancement. Firstly, the traditional atmospheric scattering model is optimized based on the dark channel prior theory. Second, a luminance augmentation branch is proposed to improve the image brightness. Finally, the images processed by both the model optimization and luminance enhancement branches are fused and subsequently color-corrected to enhance the visual quality of the images. Experimental results on the RESIDE dataset and realworld images show that our method outperforms classical and the latest dehazing methods.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2286-3540