基于人眼对比度敏感视觉特性的图像质量评价方法  

Image Quality Assessment Method Based on Contrast Sensitivity Characteristics of Human Vision System

作　　者：姚军财[1] 

机构地区：[1]陕西理工学院物理系,陕西汉中723000

出　　处：《液晶与显示》2011年第3期390-396,共7页Chinese Journal of Liquid Crystals and Displays

基　　金：国家自然科学基金(No.10447005);陕西省教育厅专项科研基金(No.2010JK463)

摘　　要：结合传统采用的均方误差和峰值信噪比评价灰度图像质量的统计方法和色差描述彩色图像差异的特点,提出了一种基于人眼视觉特性的色差均方误差和色差峰值信噪比统计参数的评价彩色图像质量的方法,并通过实验进行了验证。实验结果与德克萨斯大学奥斯汀分校的LIVE图像库中的图像主观评价结果基本符合,表明所提出的方法不仅从统计意义上有效地反映了局部甚至像点间的差别,而且由于该统计方法同时考虑了人眼视觉特性,提高了客观评价与主观评价的一致性。Combined the traditional gray image quality assessment method described by the mean square error and peak signal to noise ratio and color differences to describe color image difference,a color image quality assessment method described by the color difference MSE and color difference PSNR statistical parameters based on human vision system is proposed,which is validated by the experiments.Experimental results match the subjective image quality assessment provided by LIVE subjective image library,University of Texas at Austin very much,which shows that the method not only effectively reflects the difference among the local location or points from the statistical significance,and at the same time,which improves the consistency between the objective evaluation and subjective evaluation because the statistical method takes into account the human vision system characteristics.

关 键 词：图像质量评价 人眼视觉特性 峰值信噪比 色差 

分 类 号：TP309[自动化与计算机技术—计算机系统结构] TP391[自动化与计算机技术—计算机科学与技术]