基于暗原色先验的图像去尘算法

Under the influence of the coal mine working environment,collected image clarity is low and it effect is not good. Aiming at the problem of image degradation, in this paper uses the improved dust removing algorithm based on dark channel prior improvement. The algorithm uses LMS human visual model convert RGB three-channel to new transmittance three-channel color space,and the correlation between each channel are lower than those of RGB. Experimental results show that this color space can more accurate reconstruction image based on dark channel prior, and the result of image to dust is satisfactory.