Abstract: The opacity of the media in the atmosphere（such as fog, smoke and water etc.） images are the outdoor scene degraded, and dark channel prior method has achieved good results in the treatment of defogging, but this method consumes large memory and the longer image processing time. To solve this problem, this paper proposes a fast image dehazing method for image defogging based on the dark channel prior, which uses bilateral filtering and corrosion. Firstly, this method uses dark channel prior to get rough soft matting, then uses corosion processing for the boundary point; at last, bilateral filter is used to get the detailed soft matting and achieve the final restored image. Experimental results show that this method not only reaches the requirement of image restoration, but also the processing speed of this method is much better than the original one, and it will be used in video haze.

Key words: dehazing, dark channel prior, bilateral filter, corrosion

摘要： 大气中混浊媒介（如雾、烟、水滴等）的存在致使户外场景图像降质，暗原色先验法对单幅图像去雾具有较好的效果，但是在处理图像时存在存储消耗大、耗时长等不足。针对这一不足，在暗原色先验原理基础上，提出一种融合双边滤波和腐蚀处理的图像快速去雾方法。该方法利用暗原色先验得到粗略软抠图，通过腐蚀处理粗略软抠图的边界点，再利用高斯型双边滤波对腐蚀后的透射图进行细化操作得到最终复原图像。实验结果表明该方法不仅达到了图像复原的要求，且比原方法在处理速度上有了较大的提高，同时还可以将其进一步应用于视频图像去雾。

关键词: 图像去雾, 暗原色先验, 双边滤波, 腐蚀