Images captured in foggy weather often suffer from bad visibility, because of bad atmospheric visibility. The existence of particles, dust and water impurities always fade the color and lead to a bad contrast of the observed objects. And then, we will feel indistinct by our eyes. We introduce a novel algorithm of single image that based on hybrid filter in this paper. The algorithm is based on an advanced median filter, and combined with average filter. The visibility of photos has been greatly improved by use of iteration. Experimental results on a variety of haze images demonstrate the effectiveness of the advanced method, which contains the fast speed of Median filter and also gets more distinct in details compared with other algorithms. The main advantage of the proposed algorithm compared with others is its speed. This speed allows visibility restoration to be applied for the first time within real-time processing applications such as sign, lane-marking and obstacle detection.