Computer Engineering and Applications ›› 2010, Vol. 46 ›› Issue (7): 162-164.DOI: 10.3778/j.issn.1002-8331.2010.07.049

• 图形、图像、模式识别 • Previous Articles     Next Articles

Weighted adaptive face recognition using singular value and feature-matrix

WANG Hong-yong,LIAO Hai-bin,DUAN Xin-hua,DING Mi   

  1. College of Information Science and Engineering,Henan University of Technology,Zhengzhou 450001,China
  • Received:2008-09-12 Revised:2008-12-11 Online:2010-03-01 Published:2010-03-01
  • Contact: WANG Hong-yong

基于奇异值与特征融合矩阵的自适应人脸识别

王宏勇,廖海斌,段新华,丁 汨   

  1. 河南工业大学 信息科学与工程学院,郑州 450001
  • 通讯作者: 王宏勇

Abstract: In order to recognize face quickly and accurately,a weighted adaptive algorithm based on the conjunction of partial and entire features is proposed.Global features and local features of six key parts of face are extracted respectively by singular value to get the feature-matrix,dynamic method of how to choose the weights of local features and formula of how to obtain the feature-matrix is given and proved.Finally,the traditional Support Vector Machine(SVM) is improved to recognize the unknown faces.Experiments show that the proposed algorithm can not only calculate efficiently and work easily,but also deal with low recognition rate problems in SVD and small sample problems in LDA effectively,and the result indicates that it has a good performance and future.

Key words: face recognition, feature extraction, Singular Value Decomposition(SVD), wavelet decomposition, feature-matrix

摘要: 为了准确快速地进行人脸识别,提出了融合人脸全局和局部特征的自适应算法。该方法利用奇异值提取人脸的全局特征和6个关键部分的局部特征进行加权融合得出特征融合矩阵,同时给出了局部特征权值的动态选择方法,并证明了特征融合矩阵的推导公式;最后使用改进的支持向量机(SVM)方法进行分类识别。试验表明,该方法不仅计算速度快、简单易用,而且有效解决了SVD识别率不高和LDA小样本空间问题,应用前景良好。

关键词: 人脸识别, 特征提取, 奇异值分解, 小波分解, 特征融合矩阵

CLC Number: