An Adaptive-Window-Based Image Dehazing Method

In this paper, we propose a adaptive-window-based dehazing method for hazy images. The proposed method is composed of four parts: dark channel measurement, bilateral filtering, atmosphere light and transmission map estimation, and scene radiance recovery. Dissimilar to existing methods, we proposed an adaptive-window-based algorithm to obtain the dark channel. Experimental results demonstrate that the proposed adaptive-window-based method can reduce blocking artifacts in dehazing images and maintain the visual quality of dehazed images high. In addition, the proposed method is faster than the existing method [2].