景深边缘优化的暗通道去雾算法

雾天条件下拍摄的图像存在低对比度和低场景可见度的问题,传统的去雾算法容易产生 Halo 效应.为此,提出了景深边缘优化的暗通道去雾算法.该算法是基于暗原色先验理论,需要考虑消除景深边缘处的白色晕块,方法是对图像景深边缘和非景深边缘分别采用不同的模板处理得到透射图.引导滤波修复透射图,在去除多余的纹理信息的同时增强了雾天图像的边缘信息.并采用新增可见边之比、梯度化均值对去雾图像进行客观评价.实验表明该算法不仅能够有效地去除图像中的雾气,而且有效地降低了Halo效应,从整体上改善了去雾复原图像的视觉效果.

Dark channel prior dehazing algorithm based on depth edges optimization

A foggy image has low contrast and low visibility.Traditional dehazing algorithms suffer from the halo phenomenon.So I proposed a new dark channel prior algorithm based depth-edge.When a fast haze-removal algo-rithm based on the theory of dark channel prior,it should take the elimination of white halo on the depth-edges.We estimate the transmission map through processing the depth-edges and non depth-edges respectively with different neighborhoods.Guide filtering is adopted to modify the transmittance.It removes redundant texture information and brings the depth edge information.Compared by rate of new visible edges and mean ration of the gradient at visible edges.Presented method processes the fog image effectively the visual effect of the haze-free image and it eliminates the halo.The visual effect of whole defogging restored image could be improved.