Clearing research on fog and dust images in coalmine intelligent video surveillance.

There are many fog and dust images with much random noise in coalmine intelligent video surveillance. Therefore, the image degradation has seriously affected the subsequent video image processing. In this paper, an algorithm based on dark channel prior and bilateral filter was proposed which can realize fog and noise simultaneously removing. By combining the atmospheric scattering model, the degradation model of fog and dust images was established. Considering the characteristics of fog images,the methods and procedures for estimating the air light and rough transmittance were designed using the dark channel prior. By analyzing the optimization requirement of the rough transmission map and the bilateral filter characteristic, a joint bilateral filter was introduced for quickly obtaining the fine transmission map. The regularization objective function was constructed on the image degradation model. By solving a converted image and Gaussian bilateral filtering the image, fog and dust removal and simultaneously denoising were realized. Experimental results verify that the proposed algorithm is effective which computational efficiency is greatly improved compared with various restoration algorithms. Because of good restoring quality the algorithm is suitable for the environment of coalmine intelligent video surveillance. [ABSTRACT FROM AUTHOR]