J4 ›› 2012, Vol. 50 ›› Issue (02): 293-298.

• 计算机科学 • 上一篇    下一篇

基于模糊聚类分析和多特征融合的人脸识别方法

高璐, 李文辉, 王莹   

  1. 吉林大学 计算机科学与技术学院, 长春 130012
  • 收稿日期:2011-01-23 出版日期:2012-03-26 发布日期:2012-03-21
  • 通讯作者: 李文辉 E-mail:liwh@jlu.edu.cn

A Face Recognition Method Based on FuzzyClustering Analysis and Multifeature Fusion

GAO Lu, LI Wenhui, WANG Ying   

  1. College of Computer Science and Technology, Jilin University, Changchun 130012, China
  • Received:2011-01-23 Online:2012-03-26 Published:2012-03-21
  • Contact: LI Wenhui E-mail:liwh@jlu.edu.cn

摘要:

提出一种使用模糊聚类融合线性子空间特征和Gabor小波特征的人脸识别方法. 通过分析样本在子空间的聚类情况, 对处于不同聚类边界的样本采用Gabor特征进行二次识别. 实验结果表明, 该方法能在保证识别率的前提下, 有效减少平均识别时间.

关键词: 计算机应用, 人脸识别, 特征融合, 线性子空间, 模糊聚类分析, Gabor小波

Abstract:

A method of using fuzzy clustering analysis to combine linear subspace features and Gabor features to recognize faces was proposed. Firstly it analyzes the distribution of sample database in linear space. Then it uses Gabor features to recognize faces that locate in boundary of different cluster centers. Experimental results show the presented method has high recognition rate, meanwhile it reduces average recognition time.

Key words: computer application; face recognition; feature fusion; linear subspace; fuzzy clustering analysis; Gabor wavelet

中图分类号: 

  • TP391