Research on Image Defogging Algorithms Based on Color Attenuation Prior

Abstract

As for the defogging algorithm of color attenuation prior, an optimization algorithm based on color attenuation prior is proposed to solve the problem of image blurring in high frequency area of close-up view. Firstly, by improving the scene depth estimation information, the nonlinear model about scene depth is re-established to find the appropriate scene depth. Second, binary tree search algorithm is used to find the optimal atmospheric light value, and finally, a certain range of parameters are adjusted to achieve model optimization. The experimental results show that the algorithm solves the problem of blurring the close-up area and improves the speed of the algorithm.