Fast fusion-based underwater image enhancement with adaptive color correction and contrast enhancement

Xinzhe Yao,Xiuman Liang,Haifeng Yu,Zhendong Liu

DOI: https://doi.org/10.1007/s12145-024-01620-z

2024-12-05

Earth Science Informatics

Abstract:Since the scattering of suspension particles and the absorption of light by the water, single underwater image often suffers from some serious color cast and hazing problems, which hinder the application of advance vision technology in the underwater. To address these degradation problems, we propose a fast fusion-based method for underwater image enhancement. First, an adaptive color correction module with color cast judgment is designed to adjust the color cast of different scenes. Then, we design the dehazed and detail enhancement module to adjust the luminance channel of the image. Finally, a Laplace decomposition and multi-scale fusion strategy based on luminance channel is proposed to enhance the comprehensive contrast of the image. Our method is not dependent on complex physical imaging models while processing only at the channel level, which reduces the running time of the algorithm. The experimental results demonstrate that our algorithm outperforms the state-of-the-art algorithms of the same field. Besides, our method is equally applicable to the images of foggy and sandstorm.

geosciences, multidisciplinary,computer science, interdisciplinary applications