Image dehazing using regularized optimization

The presence of haze shifts the color and degrades the visibility of outdoor scenes in digital images. In this paper, we propose a novel and effective optimization algorithm for single image dehazing. We first formulate the dehazing model into a linear convex optimization problem, and we develop its cost function based on two basic observations: first, a hazy image exhibits low contrast in general; second, the distance-map from the scene to the camera, is piecewise smooth. Then, we implement specific algorithm for our optimization problem using Split Bregman iteration. The experimental results show that our proposed algorithm not only enhances the contrast but also preserves the details and sharp edges. Our results demonstrate the effectiveness of the proposed optimization algorithm for dehazing