视频质量评价方法综述  

A Review on the Video Quality Assessment Methods

作　　者：佟雨兵[1] 胡薇薇[1] 杨东凯[1] 张其善[1] 

机构地区：[1]北京航空航天大学电子信息工程学院,北京100083

出　　处：《计算机辅助设计与图形学学报》2006年第5期735-741,共7页Journal of Computer-Aided Design & Computer Graphics

摘　　要：论述了视频质量的主观和客观评价方法,重点阐述了视频质量的客观评价方法.对目前常用的视频质量客观评价指标峰值信噪比和均方误差进行了分析,并指出其存在的问题;将客观评价方法归结为基于人眼视觉系统生理特征和基于结构失真的两大类,并结合仿真实验予以评述;最后对视频质量评价技术的发展进行了预测和展望.In this work, objective and subjective video quality assessment methods are reviewed and the emphasis is put on the objective one. PSNR （peak signal noise ratio） and MSE （mean square error）, the two commonly used assessment methods, are analyzed and their shortcomings are figured out. In addition,the objective video quality assessment is divided into the two broad categories, namely the human vision based one and the structure distortion based one, and simulation results are reported to show the performance of typical algorithms. Finally, we also list some predictions on the future trend and possible promising issues.

关 键 词：视频质量评价 人眼视觉系统 结构失真 

分 类 号：TP918.8[自动化与计算机技术]