计算机工程 ›› 2011, Vol. 37 ›› Issue (17): 19-22,28.doi: 10.3969/j.issn.1000-3428.2011.17.005

• 专栏 • 上一篇    下一篇

基于KPCA和FMSD的人脸识别

曾接贤,田金权,符 祥   

  1. (南昌航空大学软件学院,南昌 330063)
  • 收稿日期:2011-04-01 出版日期:2011-09-05 发布日期:2011-09-05
  • 作者简介:曾接贤(1958-),男,教授,主研方向:图像处理,计算机视觉;田金权,硕士研究生;符 祥,博士
  • 基金项目:

    国家自然科学基金资助项目(60675022);江西省自然科学基金资助项目(2008GZS0034);航空科学基金资助项目(200855560 17)

Face Recognition Based on Kernel Principal Component Analysis and Fuzzy Maximum Scatter Difference

ZENG Jie-xian, TIAN Jin-quan, FU Xiang   

  1. (School of Software, Nanchang Hangkong University, Nanchang 330063, China)
  • Received:2011-04-01 Online:2011-09-05 Published:2011-09-05

摘要:

针对核最大散度差(KMSD)方法在人脸识别中存在边缘类和次优性问题,提出一种基于核主成分分析(KPCA)与模糊最大散度差(FMSD)的人脸识别方法(KFMSD)。利用KPCA方法提取人脸的非线性结构特征,选取投影后类间散度大于类内散度的特征向量作为最优投影轴,采用FMSD方法,根据隶属度函数将样本的原始分布信息完全融入人脸的特征提取中,采用最近邻分类器进行分类识别。在ORL和YALE人脸库上的实验证明了KFMSD方法的有效性。

关键词: 人脸识别, 核主成分分析, 模糊最大散度差, 核最大散度差, 特征提取

Abstract:

Considering the outer classes and inferior problem in Kernel Maximum Scatter Difference(KMSD) method, a new method of face recognition based on Kernel Principal Component Analysis(KPCA) and Fuzzy Maximum Scatter Difference(FMSD) is developed. The KPCA can be benefit to develop the nonlinear structures features in faces. Selecting the eigenvectors that between-class scatter is greater than within-class scatter after projection as optimal projection axis. Distribution information of samples is represented with fuzzy membership degree in the FMSD. It uses the nearest neighbor classifier for face recognition. Experimental results on ORL and YALE face databases show the KFMSD is better than KMSD method.

Key words: face recognition, Kernel Principal Component Analysis(KPCA), Fuzzy Maximum Scatter Difference(FMSD), Kernel Maximum Scatter Difference(KMSD), feature extraction

中图分类号: