基于暗通道的图像去雾改进算法研究  

Research on improved algorithm of image dehazing based on dark channel

作　　者：刘天时[1] 惠霄霄 

机构地区：[1]西安石油大学计算机学院,西安710065

出　　处：《智能计算机与应用》2018年第1期36-39,45,共5页Intelligent Computer and Applications

基　　金：陕西省工业科技攻关项目(2015GY026;2016GY132)

摘　　要：针对现有去雾算法对图像中天空部分处理过度增强的问题,本文对基于暗通道原理的图像去雾算法进行了改进。该算法首先对图像进行基于四叉树的搜索,查找出满足条件的大气光值。其次,根据大气物理模型和暗通道原理,求出满足条件的透射率,并对其优化,以此得到去雾图像。实验结果表明,该算法不仅去雾效果好,且有效克服了Kaiming He算法对图像中天空处理不佳的缺陷,同时具有较好的实时性,在图像去雾中具有一定的应用价值。Aiming at the problem that the existing dehazing algorithm has overly enhanced the processing of the sky in the image,this paper improves the image dehazing algorithm based on the principle of dark channel. First, the algorithm searches the imagebased on quad tree, find the atmospheric light that meets the conditions. second, according to the atmospheric physical model andthe dark channel principle, find the transmission that satisfies the condition, and optimize it, thereby a dehazed image is obtained.The experimental results show that the algorithm not only has good dehazing effect, but also effectively overcomes the shortcomingsof Kaiming He algorithm which has poor handling for image contains sky. Meanwhile, the real-time ability of the algorithm isbetter. So it has certain application value in the field of image dehazing.

关 键 词：图像去雾 图像复原 大气物理模型 暗通道 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]