基于半反图像的透射率优化降雾算法

传统的暗原色先验图像降雾算法在处理不满足暗原色先验假设的明亮区域时,估计的透射率不准确。从而导致降雾后的图像色彩出现较大偏差。针对这一不足,本文提出了一种基于半反图像的透射率优化降雾算法。该算法通过明亮区域检测来获取大气光,然后用自定义函数对图像中明亮区域透射率进行修正,最后利用引导滤波器优化初始透射率,恢复出清晰的降雾图像。实验结果表明,该算法可以有效地处理图像中不满足暗原色先验假设的明亮区域,提高了户外视觉系统的鲁棒性。

Dehazing algorithm based on semi-inverse image transmission optimization

Conventional single image dehazing algorithms based on dark channel prior often fail in estimating accurate transmission maps, when the observed images contain regions with bright area, which the dark channel prior cannot deal with. Accordingly, large deviation in terms of color may appear in the dehazed results. In order to solve the problems mentioned above, in this paper, we propose a novel single image dehazing algorithm based on semi-inverse image transmission optimization. In this approach, atmospheric light is firstly obtained by detecting bright regions. Coarse transmission map of bright patches are then corrected using a self-defined function. Finally, a haze-free image is generated with the guided image filter refined transmission map. Experimental results on a variety of hazy images demonstrate the effectiveness and the robustness of the proposed method.