首页 | 本学科首页   官方微博 | 高级检索  
     

基于小波分析和KPCA的人脸识别
引用本文:杨国亮,任金霞,刘细平.基于小波分析和KPCA的人脸识别[J].自动化技术与应用,2003,22(9):29-32.
作者姓名:杨国亮  任金霞  刘细平
作者单位:南方冶金学院,机电学院,江西,赣州,341000
基金项目:南方冶金学院自选课题资助
摘    要:本文探讨了基于核函数的主成分分析方法在人脸识别中的应用,首先对人脸进行haar小波分析,得到对应的人脸小波系数,再通过计算其内积核函数实现从低维空间到高维空间的非线性映射,对高维数据进行主成分分析得到用于分类的主成分,最后采用支持向量机进行分类,实验结果表明,该方法具有良好的分类性能和鲁俸性。

关 键 词:人脸识别  小波变换  核函数主成分分忻  支持向量机
文章编号:1003-7241(2003)09-0029-04

Face Recognition Based On Wavelet Analysis And KPCA
YANG Guo-liang,REN Jin-xia,LIU Xi-ping.Face Recognition Based On Wavelet Analysis And KPCA[J].Techniques of Automation and Applications,2003,22(9):29-32.
Authors:YANG Guo-liang  REN Jin-xia  LIU Xi-ping
Abstract:This paper describes an application of Kernel Principal Component Analysis in face recognition.First,Haar wavelet transform was used to process the face image,resulting in corresponding wavelet coefficients.By using kernel functions of the integral operator, the nonlinear principal components in high dimensional feature spaces related to input spaces by some nonlinear map were computed.Final,we adopt Support Vector Machnie to classify in ORL data base.Experiment results indicate that this method has good performance and robustness.
Keywords:Face recognition  Wavelet transform  Kernel principal component analysis  Support vetor machine
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号