基于Retinex的视频自适应增强算法  

Retinex-based self-adapting enhancement algorithm for videos

作　　者：侯天峰[1] 程和生[2] 张燕[1] 

机构地区：[1]南京大学金陵学院信息科学与工程学院,江苏南京210000 [2]合肥师范学院计算机学院,安徽合肥230601

出　　处：《现代电子技术》2015年第17期68-71,74,共5页Modern Electronics Technique

摘　　要：Retinex算法是图像增强的常用方法。基于Retinex理论提出一种新的视频光照增强算法,结合图像全局及局部灰度特性引入补偿系数,增强暗区域光照的同时克服了传统算法处理后的人工痕迹,使处理结果更加自然;同时,算法用引导滤波估计照度图像,在避免光晕现象的同时可以加快算法处理速度。实验结果表明,该方法简单有效,视觉效果提升显著。Retinex algorithm is a common image enhancement method. Based on Retinex theory,a new video illumination enhancement algorithm is proposed,in which the compensation factor is introduced in combination with global and local grayscale characteristics of images. It can enhance the illumination in dark areas and overcome the artificial marks on images processed by traditional algorithms,which can make the processing results more natural. The guided filtering is applied to estimation illumination image by the algorithm,which can avoid halo phenomenon and accelerate the processing speed of the algorithm. The experimental results show that this method is simple and effective,with which the visual effect is promoted obviously.

关 键 词：RETINEX 图像增强 光照补偿 引导滤波 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]