采用亮通道先验的低照度图像增强算法  

Low Light Image Enhancement Algorithm Using Bright Channel Prior

作　　者：贾存坤 戴声奎[1,2] 卫志敏 JIA Cunkun;DAI Shengkui;WEI Zhimin(School of Information Science and Engineering,Huaqiao University,Xiamen 361021,China;Xiamen Key Laboratory of Mobile Multimedia Communication,Xiamen 361021,China)

机构地区：[1]华侨大学信息科学与工程学院,福建厦门361021 [2]厦门市移动多媒体通信重点实验室,福建厦门361021

出　　处：《华侨大学学报（自然科学版）》2018年第4期595-599,共5页Journal of Huaqiao University(Natural Science)

基　　金：福建省科技计划重点项目(2013H0030);中央高校基本科研业务费专项基金资助项目(JB-ZR1145);华侨大学研究生科研创新能力培育计划资助项目(1400401015)

摘　　要：针对低照度彩色图像亮度偏低、对比度差等问题,提出基于亮通道先验的低照度图像增强算法.首先,分析Retinex算法所存在的缺陷,提出了亮通道先验.然后,将原RGB彩色图像转换到HSV彩色空间,对亮度分量V使用亮通道先验和引导滤波估计光照分量和反射分量,并且采用自适应对数校正对光照分量进行提升.最后,将增强后的图像转换到RGB彩色空间.实验结果表明:该算法快速有效,能够很好地提升图像整体亮度和对比度,图像细节得到增强,克服了颜色失真和光晕等问题,增强后的彩色图像更为明亮、自然.In order to improve the brightness and contrast of low illumination image,a low light image enhancement algorithm based on bright channel prior is proposed in this paper.Firstly,the proposed algorithm comes up with bright channel prior in order to deal with flaw of the Retinex algorithms.Then the original RGB color image is converted to HSV color space,and the illumination image of Vcomponent is evaluated by using bright channel and guided filter,as well as the contrast of the illumination image is enhanced by adaptive logarithm correction.Finally,the enhanced image is converted to RGB color space.The experimental results show that the proposed algorithm is fast and effective,the overall image brightness and contrast is increased,as well as the image detail.Furthermore,the proposed algorithm overcome the color distortion and halos,and the enhanced color image is more bright and beautiful.

关 键 词：图像增强算法 亮通道先验 引导滤波 自适应对数校正 HSV彩色空间 RETINEX理论 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]