计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (1): 194-195.DOI: 10.3778/j.issn.1002-8331.2010.01.057

• 图形、图像、模式识别 • 上一篇    下一篇

用于人脸识别的两类主成分分析融合

杨 军1,2,张秀琼2,高志升2,袁红照2   

  1. 1.四川师范大学 计算机科学学院,成都 610066
    2.四川大学 图形图像研究所,成都 610064
  • 收稿日期:2008-07-21 修回日期:2008-08-25 出版日期:2010-01-01 发布日期:2010-01-01
  • 通讯作者: 杨 军

Fusion of two different PCAs for face recognition

YANG Jun1,2,ZHANG Xiu-qiong2,GAO Zhi-sheng2,YUAN Hong-zhao2   

  1. 1.College of Computer Science,Sichuan Normal University,Chengdu 610066,China
    2.Institute of Image & Graphic,Sichuan University,Chengdu 610064,China
  • Received:2008-07-21 Revised:2008-08-25 Online:2010-01-01 Published:2010-01-01
  • Contact: YANG Jun

摘要: 分析了基于总体离散度矩阵和总类间离散度矩阵的主成分分析的原理。利用两种方法分别提取人脸特征并进行识别。对两种方法获得的结果进行了特征层融合和决策层融合,基于ORL人脸数据库的实验表明该方法的识别性能优于单一的主成分分析方法。

关键词: 人脸识别, 主成分分析, 总体离散度矩阵, 类间离散度矩阵, 数据融合

Abstract: The principle of two different PCAs,PCA based on global scatter matrix and PCA based on global between-class scatter matrix is analyzed firstly.Two different fusion methods,feature level fusion and decision level fusion are proposed using the feature got from two different PCAs.The experiment result is displayed and data fusion method is proved to be efficient for getting better recognition rate.

Key words: face recognition, Principal Component Analysis(PCA), global scatter matrix, between-class scatter matrix, data fusion

中图分类号: