首页 | 本学科首页   官方微博 | 高级检索  
     

一种基于视觉感兴趣区域的彩色图像增强方法
引用本文:王晓红,章婷.一种基于视觉感兴趣区域的彩色图像增强方法[J].包装工程,2014,35(3):84-87,147.
作者姓名:王晓红  章婷
作者单位:上海理工大学, 上海 200093;上海理工大学, 上海 200093
摘    要:目的把视觉感兴趣区域概念引入直方图的建造中,提出了一种新的基于视觉感兴趣区域的图像增强方法,使增强效果更符合人眼视觉感知。方法首先在标准观测环境下利用先进的眼动仪设备获得人眼感兴趣区域,然后计算各子区域的平均显著值,以确定各个子区域的权重系数,最后采取类似直方图均衡化的思想,优化配比灰度级的动态范围。结果通过主客观实验结果表明,增强后的图像更符合人眼视觉感知。结论结合了视觉感知特性的直方图增强方法,弥补了传统直方图与人眼视觉感知不一致的弊端,其增强效果更佳。

关 键 词:图像增强  视觉感兴趣区域  基于视觉感知的灰度级方图  直方图均衡化
收稿时间:2013/11/8 0:00:00
修稿时间:1/1/2014 12:00:00 AM

Color Image Enhancement Based on Visual Region of Interest
WANG Xiao-hong and ZHANG Ting.Color Image Enhancement Based on Visual Region of Interest[J].Packaging Engineering,2014,35(3):84-87,147.
Authors:WANG Xiao-hong and ZHANG Ting
Affiliation:University of Shanghai for Science and Technology, Shanghai 200093, China;University of Shanghai for Science and Technology, Shanghai 200093, China
Abstract:Objective The concept of visual region of interest was brought into the construction of the histogram. We proposed a new visual image enhancement method based on region of interest, in order to make the enhanced images more in line with the visual perception of the human eye. Methods In this method, advanced eye tracker device was used in the standard observation environment and visual region of interest, namely the human saliency map was obtained, whereby the image of the significant regions was analyzed. In each sub-region, salient value was obtained and normalized, so that weighted coefficient was obtained for each sub-region. Finally a similar idea to histogram equalization was adopted to optimize the dynamic range of the ratio of gray levels. Results Through subjective and objective experiments, the experimental results showed that the enhanced images were more in line with the visual perception of the human eye. Conclusion Histogram enhancement method combined with perceptual characteristics could compensate the drawback of the traditional histogram in the inconsistency with the visual perception, resulting in more enhanced image.
Keywords:image enhancement  visual region of interest  gray levels histogram based on visual perception  histogram equalization
本文献已被 CNKI 等数据库收录!
点击此处可从《包装工程》浏览原始摘要信息
点击此处可从《包装工程》下载全文
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