单幅图像自动去雾算法

为了改善雾天环境下退化图像的视觉效果,提出一种基于物理模型的快速图像去雾算法.算法从大气散射模型出发,从有雾图像中利用腐蚀和膨胀粗略估计出大气耗散函数,再利用指导图像滤波方法细化估计大气耗散函数,进而恢复场景反照率.实验结果表明,算法可以获得更精确的大气耗散函数,复原图像的边缘轮廓及景物特征都比较清楚,可有效抑制晕环效应,且算法速度也有显著提高,可用于实时雾天图像处理.

Automatic Dehazing Algorithm of Single Image

In order to improve the visual quality of fog-degraded images, this paper proposes a fast image dehazing algorithm based on physical model. The algorithm starts from the atmospheric scattering model, and uses erosion and dilation to infer the atmospheric veil from the fog-degraded image. Then, the rough es- timate of the atmospheric veil is refined by guided-image filtering. Finally, the scene albedo is recovered by inverting this simplified model. The experimental results show that the proposed algorithm can achieve more accurate atmospheric veil, and the edge outline and scenery features of the recovered image are clear, and the halo effect is also inhibited, and the speed of the algorithm is significantly improved, which can be used in real-time fog image processing.