›› 2016, Vol. 29 ›› Issue (2): 74-.

• 论文 • 上一篇    下一篇

基于GPCA的KNNY与SVM融合的人脸识别方法

焦淑红,孙志帅   

  1. (哈尔滨工程大学 信息与通信工程学院,黑龙江 哈尔滨 150001)
  • 出版日期:2016-02-15 发布日期:2016-02-25
  • 作者简介:焦淑红(1966—),女,教授,博士生导师。研究方向:宽带信号检测等。孙志帅(1988—),男,硕士研究生。研究方向:人脸识别。

Face Recognition Based GPCA of KNN and SVM Fusion

JIAO Shuhong,SUN Zhishuai   

  1. (School of Information and Communications Engineering,Harbin Engineering University,Harbin 150001,China)
  • Online:2016-02-15 Published:2016-02-25

摘要:

针对K近邻和支持向量机人脸识别率较低的问题,采用一种KNN和SVM融合的识别方法。提出了一种Gabor小波和主成分分析进行人脸特征提取,KNN-SVM进行分类的人脸识别方法。基于ORL和YALE人脸库中进行实验,结果表明该算法较KNN和SVM中任何一个的识别率都要高,且识别率最高可达到98.89%。

关键词: K近邻, 支持向量机, Gabor小波, PCA, 人脸识别

Abstract:

In view of the poor face recognition rate of the K Nearest neighbor and support vector machine (SVM),a KNN and SVM fusion recognition method is proposed with a Gabor wavelet and principal component analysis (PCA) for face feature extraction and KNN-SVM classification method for face recognition.Experiments based on ORL and YALE face database show that the proposed algorithm offers a recognition rate up to 98.89%,higher than both KNN and SVM.

Key words: KNN;SVM;Gabor wavelet;PCA;face recognition

中图分类号: 

  • TP391.41