Single Image Dehazing Based on Sky Area Segmentation and Image Fusion

Since the dark channel prior (DCP)-based dehazing method is ineffective in sky area and will cause problem of too color distortion image, we propose a novel based on segmentation image fusion. We first segment according to characteristics non-sky then estimate atmospheric light transmission map DCP correct them, fuse original after contrast adaptive histogram equalization improve details information image. Experiments illustrate that our performs well can reduce distortion.