A Fast Image/Video Dehazing Algorithm Based on Modified Atmospheric Veil

Abstract

In this paper, we propose a fast image/video dehazing algorithm based on modified atmospheric veil. For a single image, the coarse atmospheric veil is quickly estimated by the minimum and maximum channels. We take the depth image estimated by the color attenuation prior as the guided image to refine the coarse atmospheric veil and use down sampling technique to speed up the process. In addition, a fast Fourier enhancement method is used to improve the high frequency details. Based on the image dehazing, we design an atmospheric light adjustment strategy and a transmission optimized framework for video dehazing, which can effectively reduce flickering artifacts. Experimental results demonstrate that the proposed algorithm can achieve a very good dehazing effect while keeping a comparable processing speed with state-of-the-art algorithms.