Novel single hazy image restoration method based on nonlocal total variation regularization optimization

Based on the property that the scene radiance is of high contrast and the atmospheric veil is locally smooth, a novel single hazy image restoration method based on nonlocal total variation regularization optimization is proposed in this paper. In order to obtain the atmospheric veil of a hazy image, a constrained nonlocal total variation regularization is firstly applied. Then, the accurate atmospheric veil is estimated using a nonlocal Rudin-Osher-Fatemi model, which is solved by a modified split Bregman method. Experimental results demonstrate that the proposed approach is capable of recovering the scene radiance from a single hazy image effectively, especially for the regions with multi-texture.