首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于色彩恒常理论的图像去雾技术
引用本文:芮义斌,李鹏,孙锦涛.基于色彩恒常理论的图像去雾技术[J].南京理工大学学报(自然科学版),2006,30(5):622-625.
作者姓名:芮义斌  李鹏  孙锦涛
作者单位:南京理工大学,电子工程与光电技术学院,江苏,南京,210094
基金项目:南京理工大学校科研和教改项目
摘    要:针对户外图像采集处理系统,讨论了1种有效的图像去雾方法。由于雾退化图像对比度较低,从图像增强角度入手,将雾的退化作用等效成景物的照度变化,根据色彩恒常理论,采用Frankle—McCann Retinex(FMR)算法进行图像去雾处理;结合人眼对亮度的指数视觉特性,采用软复位算法对其进行改进。仿真分析表明,改进后的算法可以有效增强雾退化图像的对比度,获得满意的视觉效果。

关 键 词:色彩恒常  图像去雾  Frankle-McCann  Retinex算法  对比度
文章编号:1005-9830(2006)05-0622-04
收稿时间:2005-07-01
修稿时间:2006-06-18

Images Defogging Techniques Based on Color Constancy Theory
RUI Yi-bin,LI Peng,SUN Jin-tao.Images Defogging Techniques Based on Color Constancy Theory[J].Journal of Nanjing University of Science and Technology(Nature Science),2006,30(5):622-625.
Authors:RUI Yi-bin  LI Peng  SUN Jin-tao
Abstract:A method of defogging images is introduced for outdoor images acquisition and processing system. The fog degraded images have low contrast. The enhancement of images is considered, and the fog degradation is assumed as illumination variation. Based on the color constancy theory, the Frankle-McCann Retinex(FMR) algorithm is deployed to remove fogging effects from color images. Considering the exponential characteristics of eyes for illumination, the FMR algorithm is amended with soft resetting. The experimental results demonstrate the satisfying visual effects of this approach.
Keywords:color constancy  images defogging  Frankle-McCann Retinex algorithm  contrast
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号