In this paper, a single image defogging method based on dark channel is proposed to prevent color distortion in the defogging image. One of the drawbacks of the conventional DCP-based defogging methods is low-contrast quality for the defogging image. Various enhancement skills can be applied to the defogging image in order to get the fidelity and to improve the drawback referred above, while they can cause some problems such as a color distortion or a saturation distortion.
In order to solve these problems, in this paper, a histogram information with DCP on transmission estimation regions and a hue weight are used in cost function. The local histogram information can reduce the low-contrast problem on the defogging image, and the hue weight in HSI color space can control the ratio of cost function between DCP cost term and a histogram cost term. And also, a saturation weight in HSI applied to the original fog image is used in order to prevent the saturation distortion. The validity of the proposed method is assured by way of experimental results for several fog images and it is concluded that the proposed algorithm is more efficient than the conventional defogging algorithms.