VRHI: Visibility Restoration for Hazy Images Using a Haze Density Model

In this paper, a image processing method called VRHI is developed to enhance single hazy images. More specifically, inspired by visual characteristics of haze, a haze density estimation model is designed to predict the haze distribution. According to this recognized haze distribution, a quadtree based recursive strategy is subsequently proposed to locate the atmospheric light. Finally, by combining a global-wise adjusting mechanism and atmospheric scattering model, the haze cover in an image can be easily excluded using the estimated parameters. It is worth mentioning that VRHI is based on whole image to search the unknown parameters, thereby avoiding some unfavorable phenomena, e.g., over-enhancement and color distortion. Extensive experiments on real-world images and well-known dehazing datasets show that VRHI outperforms state-of-the-art techniques in robustness and effectiveness.