AUTHOR=Zhang Hao , Gong Longxiang , Li Xiangchun , Liu Fei , Yin Jiawei 

TITLE=An underwater imaging method of enhancement via multi-scale weighted fusion

JOURNAL=Frontiers in Marine Science

VOLUME=Volume 10 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/marine-science/articles/10.3389/fmars.2023.1150593

DOI=10.3389/fmars.2023.1150593

ISSN=2296-7745

ABSTRACT=Blurred image quality and color distortion, resulting from light absorption and scattering impacts of water medium, are all significant issues in underwater optical imaging and are less conducive to our viewing. To improve underwater imaging quality without using an active light source, we propose a multi-scale weighted fusion method in this study. Firstly, we begin by merging two images as initials and use these original images to create secondary images for enhancement, brightness correction, and contrast enhancement. Meanwhile, we adjust their weights to minimize the effect on image fusion, especially in edge, color, and contrast. To avoid sharp image weight transition to a ghost image of the low-frequency component, we then proposed and used a multi-scale fusion method when reconstructing the images. This method fixes color distortion and improves underwater image contrast by reducingscattering and blurring impacts of water. As demonstrated by experimental results, the imaging fusion method in this paper not only effectively improves image fidelity to sharpness and color in a wide scope compared with the latest underwater imaging methods but also strengthens the visual perception of our review.