基于改进暗通道先验的交通图像去雾新方法

针对交通场景图像中由于雾霾导致的图像目标主体不清晰,影响监控效果的问题,提出一种基于导向滤波与自适应色阶调整的改进暗通道图像去雾新方法.首先,基于暗通道原理对原始图像进行映射处理,从而得到大气光成分与透射率的估计值,并利用多维导向滤波方法对大气透射率估计值进行优化处理;然后,根据图像降质过程的逆过程,求解雾霾图像清晰化处理初始结果;最后,利用多通道自适应色阶调整方法进一步优化初始结果,解决初始结果整体亮度较暗、不利于监控系统后期处理的问题.实验结果表明,清晰化处理后的图像具有较高的亮度和对比度值,较好地保留并增强了图像的边缘和细节信息,算法去雾霾效果显著,针对交通场景图像处理的自适应性较高.

Improved dehazing method for traffic images based on dark channel prior

The monitoring effect is influenced by haze which can make targets in images blurred. To resolve this problem, an improved image dehazing method based on dark channel prior by using the guided filter and adaptive color levels adjustment is proposed. Firstly, mapping operations are executed to estimate the atmospheric composition and transmissivity according to dark channel prior. The estimated value of transmissivity is optimized by using the multi-dimensional guided filter. Then, an initial dehazed result is generated according to the inverse process of image degradation. The light intensity of the initial result is usually low, which makes monitoring equipment hard to perform further actions. Thus multi-channel adaptive color levels adjustment is finally used to optimize the initial result. Experimental results show that the haze removal images have higher light intensity, contrast and better sharpness. The edge and detail features are enhanced significantly. The dehazing effect is remarkable, and the adaptability of the proposed method is high aiming at traffic images collected in hazy weather.