摘要
提出了一种基于二维小波分解和融合多特征的2DPCA(简称MMP-2DPCA)人脸识别方法.该方法对于人脸表情变化不敏感,能够很好地压缩和表征原始人脸图像;融合图像既能反映人脸的全局特征,又能反映人脸的局部特征,具有更强的表达能力和判别能力.在ORL人脸库上的实验表明:MMP-2DPCA方法具有有效性.
A new face recognition method based on multi-modal parts and Two-dimensional principal component analysis（MMP-2DPCA） is proposed.Althongh this nwe method is insensitive to the changes in face expression,however,it can express the principal features of the face image by compressing data,and can express both local and holistic features of the face images by blending images.The results of the experiments on the face databases of ORL demonstrate the effectiveness of the proposed new method and its algorithm.
出处
《中南民族大学学报（自然科学版）》
CAS
2013年第1期70-74,79,共6页
Journal of South-Central University for Nationalities：Natural Science Edition
基金
国家自然科学基金资助项目(61141010)
武汉市科技供需对接计划项目(201051824575)
武汉市科技攻关计划项目(201210421136)
中南民族大学智能无线通信湖北省重点实验室开放课题(IWC2012005)
关键词
人脸识别
主成分分析
MMP-2DPCA方法
二维小波分解
特征提取
face recognition
principal component analysis
MMP-2DPCA
two-dimensional wavelet decomposition
feature extraction
