基于自适应参数大小的交通图像去雾霾算法  

Traffic Image De-haze Algorithm Based on Adaptive Parameter Size

作　　者：李同亮[1] 肖杰[1] 翟东海[1] 

机构地区：[1]西南交通大学信息科学与技术学院,四川成都610031

出　　处：《西南科技大学学报》2014年第4期67-71,共5页Journal of Southwest University of Science and Technology

基　　金：国家自然科学基金项目(61461048);国家社会科学基金项目(12EF119);西藏自治区科技计划重点项目(Z2013B28G28/02);四川省科技创新苗子工程资助项目(20132010);西南交通大学研究生创新实验实践项目(YC201404215)

摘　　要：原始的基于暗通道先验理论的去雾霾算法对于一些户外场景图像去雾霾取得了一定的效果,但不能很好地处理交通图像中的一些白色区域,会导致该区域色彩失真;同时,在处理较高分辨率交通图像时,基于软抠图的透射率优化算法需要消耗大量的计算和存储资源。针对这两个问题,首先对交通图像的白色区域展开研究,分析了基于暗通道理论的去雾霾算法在白色区域产生色彩失真的原因,并基于此提出一种自适应参数大小的透射率求解模型;其次,在透射率优化过程中舍弃效率低下的软抠图算法,提出一种基于块的透射率优化算法。将所提算法用于实例验证,结果表明该算法不但可以保证交通图像的白色区域的色彩不失真,而且提高了去雾霾效率。Although the original cant achievements in de - hazing algorithm based on the dark channel prior theory has made some signifisome outdoor scene images, it cannot deal with some white areas of the traffic haze images. When dealing with the white areas, it will cause color distortion. And while processing high - resolution haze traffic images, the transmittance - optimized algorithm based on soft matting im- ages needs to consume a lot of computing and storage resources. To overcome these difficulties, firstly, by studying on the white area of haze traffic images and analyzing the reasons of color distortion, we propose a transmittance solving model based on adaptive parameter size; secondly, abandoning the inefficient soft matting algorithm, this paper proposes an algorithm based on block that can quickly optimize the transmittance. Experimental results show that the proposed algorithm ensures that the color distortion of the white area is eliminated while improving the efficiency of the algorithm.

关 键 词：暗通道先验 雾霾图像 时间复杂度 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]