English

应用科学学报 ›› 2010, Vol. 28 ›› Issue (3): 271-276.

• 信号与信息处理 • 上一篇    下一篇

利用二元结构特征的人脸识别

胡晓1， 俞王新2， 姚菁1   

  1. 1. 广州大学电子信息工程系，广州510006
    2. 上海交通大学生物医学工程系，上海200240
  • 收稿日期:2009-10-16 修回日期:2010-01-27 出版日期:2010-05-21 发布日期:2010-05-21
  • 作者简介:胡晓，博士，副教授，研究方向：信号检测和处理，E-mail: huxiao@gzhu.edu.cn
  • 基金资助:

    广东省教育部产学研结合项目基金(No.2009B090600034)；广东省科技计划项目基金(No.2009B060700124)；广州大学产学研培育项目基金(No.7)资助

Face Recognition Using Binary Structure-Based Feature Selection

HU Xiao1, YU Wang-xin2, YAO Jing1   

  1. 1. Department of Electronic and Information Engineering, Guangzhou University, Guangzhou 510006, China
    2. Department of Biomedical Engineering, Shanghai Jiaotong University, Shanghai 200240, China
  • Received:2009-10-16 Revised:2010-01-27 Online:2010-05-21 Published:2010-05-21

PDF

1780

被引次数

6

摘要：

提出了一种基于二元结构特征提取的人脸识别算法. 该算法将所有类进行两两组合，以两类分类器为基础，为每个两两组合类间的识别挑选最适合分类的特征构成特征选取空间. 对未知样本进行测试时，在特征选取空间中计算测试样本与所有训练类的相似度，将未知样本判断为与之相似度最大的类. 运用AT&T和AR人脸数据库对该算法进行性能测试，与其他算法相比，该算法能在较小的特征维数下获得更高的识别率.

关键词: 人脸识别, 主成分分析, 线性判别分析, 特征提取

Abstract:

This paper proposes a binary structure feature selection (BFS) for face recognition. In the proposed method, all classes are combined in pairs. Based on the two-class classifier, the most suitable features for discriminating these two classes are chosen to form a feature-selected space. During the test on an unknown sample image, similarities between the unknown image and all training classes are calculated in the featureselected space. The unknown image is thus judged to belong to the class which shows the highest similarity. Performance of the method has been tested with the AT&T and AR face databases. The results show that, compared with other methods, the proposed technique can achieve higher recognition rate with a low feature dimension.

Key words:  face recognition, principal component analysis, linear discriminant analysis, feature extraction

中图分类号: