计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (1): 72-74.

• 学术探讨 • 上一篇    下一篇

融合典型相关与最大散度差的特征抽取方法

彭倩倩1,陈才扣1,2,刘永俊1,3   

  1. 1.扬州大学 计算机科学与工程系,江苏 扬州 225009
    2.南京理工大学 计算机科学与工程系,南京 210094
    3.常熟理工学院 软件工程系,江苏 常熟 215500
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-01-01 发布日期:2008-01-01
  • 通讯作者: 彭倩倩

Combination of canonical correlation analysis and maximum scatter dfference discriminate analysis for feature extraction

PENG Qian-qian1,CHEN Cai-kou1,2,LIU Yong-jun1,3   

  1. 1.Department of Computer Science and Engineering,Yangzhou University,Yangzhou,Jiangsu 225009,China
    2.Department of Computer Science and Engineering,Nanjing University of Science and Technology,Nanjing 210094,China
    3.Department of Software Engineering,Changshu Insitute of Technology,Changshu,Jiangsu 215500,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-01-01 Published:2008-01-01
  • Contact: PENG Qian-qian

摘要: 提出了一种融合典型相关分析与最大散度差鉴别分析的特征抽取新方法。该方法首先利用典型相关分析方法实现了特征信息的融合,有效地消除了特征之间的信息冗余。然后,通过采用最大散度差鉴别分析方法将训练样本中的类别信息加以充分的利用,从而有效的提高了人脸识别的正确率。最后,在ORL标准人脸库上和Yale人脸库上的实验结果验证了本文算法的有效性。

关键词: 典型相关分析, 最大散度差鉴别分析, 人脸识别

Abstract: A new method of combination of Canonical Correlation Analysis(CCA) and Maximum Scatter Difference Discriminate Analysis(MSDDA) is developed in this paper.According to the idea of CCA,it suits for information fusion;What is more,it eliminates the redundant information within the features.According to the idea of Maximum Scatter Difference Discriminate Analysis (MSDDA),the information of classes is fully utilized so the correct rate of face recognition is increased much more.Finally,extensive experiments performed on both ORL face database and YALE face database verify the effectiveness of the proposed method.

Key words: Canonical Correlation Analysis, Maximum Scatter Difference Discriminate Analysis, face recognition