An underwater image enhancement method based on color correction and local unsharp masking

Abstract

The underwater images collected by optical cameras have different problems due to the camera equipment, underwater environment and light source, which lead to the problems of chromatic aberration and low contrast of the images collected by ordinary people for subsequent use. The reasons are that the reflection of light by impurities in the water and the absorption efficiency of different wavelengths of light under water, and under different depths of field conditions often leads to color Bias and contrast are low. To solve this problem, we adopt a method based on color correction and local unsharp masking for the fusion of underwater images. Firstly, color is improved by correcting for color shift using the red channel and gray world methods, combined with CLAHE. secondly, for the low contrast problem, the adaptive mask coefficient and the local unsharp mask are used to separate the background and foreground, which enhances the recognition of underwater target and background, and effectively improves the contrast of images. Target. Finally, the weighted fusion of Gaussian pyramid is completed by using the method of combining four weights. The experimental results show that the algorithm proposed in this paper has a strong generalization ability, and can improve the quality of images in different underwater scenes which is according to different objective evaluation indicators, the algorithm in this paper also achieves the best results.