论文:2012,Vol:30,Issue(3):451-456
引用本文:
张兆林, 史浩山, 万帅. 基于线性回归分析的视频质量评估方法[J]. 西北工业大学
Zhang Zhaolin, Shi Haoshan, Wan Shuai. A Computationally Efficient Video Quality Assessment Method Based on Linear Regression Analysis[J]. Northwestern polytechnical university

基于线性回归分析的视频质量评估方法
张兆林, 史浩山, 万帅
西北工业大学 电子信息学院,陕西 西安 710072
摘要:
文章提出了一种利用线性回归分析方法的无参考视频序列质量评估算法。该算法主要利用帧间编码帧的比特数和该帧与其参考帧的差异两个参数进行线性回归分析来评估视频质量。该方法不需要原始参考视频,算法简单。通过对标准视频序列的仿真实验,该算法可有效评价不同视频的编码质量,使用该质量评估方法测得的失真视频客观质量评分与其主观质量评分有很好的一致性。
关键词:    视频质量评估    无参考    线性回归分析   
A Computationally Efficient Video Quality Assessment Method Based on Linear Regression Analysis
Zhang Zhaolin, Shi Haoshan, Wan Shuai
Department of Electronics Engineering,Northwestern Polytechnical University,Xi'an 710072,China
Abstract:
Sections 1 and 2 of the full paper explain the assessment method mentioned in the title,which we be-lieve is more computationally efficient than existing ones. Section 1 briefs the state of the art. Section 2 is entitled"No-Reference Video Sequence Quality Assessment Model" ; it needs to be divided into four subsections. Its coreconsists of: (1) we employ two parameters: the bit rate for each inter-coded frame and the difference between theinter-coded frame and its reference frame; (2) we carry out the linear regression analysis of the two parameters toassess the quality of a video sequence. Section 3 did experiments on five standard CIF video sequences to verify theeffectiveness of our assessment method; the experimental results,given in Figs. 3 and 4 and Table 1,and their a-nalysis show preliminarily that: (1) our assessment method is indeed more computationally efficient than existingones and does not require the original video sequence for reference; (2) the assessment results obtained with ourassessment method,which falls into the category of objective quality assessment,are very close to those obtainedwith the subjective assessment method.
Key words:    algorithms    computational efficiency    linear regression    measurements    models    signal to noise ratio    no-reference    video quality assessment    video sequence   
收稿日期: 2011-06-18     修回日期:
DOI:
基金项目: 国家自然科学基金(F010204)资助
通讯作者:     Email:
作者简介: 张兆林(1978-),西北工业大学讲师、博士研究生,主要从事视频质量评估、误码掩盖及阵列信号处理的研究。
相关功能
PDF(2035KB) Free
打印本文
把本文推荐给朋友
作者相关文章
张兆林  在本刊中的所有文章
史浩山  在本刊中的所有文章
万帅  在本刊中的所有文章

参考文献:
[1] Farias Mylène C Q,Mitra Sanjit K.No-Reference Video Quality Metric Based on Artifact Measurements.IEEE InternationalConference on Image Processing,Genova,Italy,2005,141-146
[2] Zhang Zhaolin,Shi Haoshan,Wan Shuai.A Novel Blind Measurement of Blocking Artifacts for H.264/AVC Video.Proceedingsof the 5th International Conference on Image and Graphics,Xi'an,China,2010,262-265
[3] Jiang Xiuhua,Meng Fang,Xu Jiangbo,Zhou Wei.No-Reference Perceptual Video Quality Measurement for High DefinitionVideos Based on an Artificial Neural Network.Proceedings of the 2008 International Conference on Computer and Electrical En-gineering,Phuket,Thailand,2008,424-427
[4] Yamagishi K,Hayashi T.Opinion Model Using Psychological Factors for Interactive Multimodal Service.IEICE Trans on Com-munication,2006,2: 281-288
[5] Choe Jihwan,Lee Kwon,Lee Chulhee.No-Reference Video Quality Measurement Using Neural Networks.16th InternationalConference on Digital Signal Processing,2009
[6] Yang Fuzheng,Wan Shuai,Xie Qingpeng,Wu Hongren.No-Reference Quality Assessment for Networked Video via Primary A-nalysis of Bit Stream.IEEE Trans on Circuits and Systems for Video Technology,2010,20(11): 1544-1554
[7] Yao Jixian,Zhang Yuan,Xu Guizhong,Jin Meng.No-Reference Objective Quality Assessment for Video Communication Serv-ices Based on Feature Extraction.Proceedings of the 2009 2nd International Congress on Image and Signal Processing,Tianjin,China,2009
[8] VQEG Draft Final Report from the Video Quality Experts Group on the Validation of Objective Models of Video Quality Assess-ment.http: //www.its.bldrdoc.gov/vqeg