• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2012, Vol. 34 ›› Issue (10): 88-91.

• 论文 • 上一篇    下一篇

一种改进的基于奇异值扰动的单样本人脸识别方法

刘 嵩1,2,李时东1,郑明辉1   

  1. (1.湖北民族学院信息工程学院,湖北 恩施 445000;2.华中师范大学物理科学与技术学院,湖北 武汉 430079)
  • 收稿日期:2012-04-25 修回日期:2012-07-10 出版日期:2012-10-25 发布日期:2012-10-25
  • 基金资助:

    湖北省自然科学基金资助项目(2009CDB069,2011CDC017);湖北省教育厅创新团队资助项目(T201214)

An Improved Face Recognition Method Based on SingularValuePerturbed with Single Image

LIU Song1,2,LI Shidong1,ZHENG Minghui1   

  1. (1.College of Information Engineering,Hubei University for Nationalities,Enshi 445000;2.College of Physical Science and Technology,Huazhong Normal University,Wuhan 430079,China)
  • Received:2012-04-25 Revised:2012-07-10 Online:2012-10-25 Published:2012-10-25

摘要:

针对单训练样本情况下人脸识别性能不佳的问题,本文提出了一种改进的基于奇异值扰动的人脸识别方法。首先通过奇异值扰动方法扩展人脸样本,然后运用小波变换压缩扩展样本,选择小波变换分解后的低频分量作为子图像,再采用核主成分分析提取人脸的高阶特征,最后根据最近邻分类器分类。在ORL和Yale数据库上的仿真实验证明了本文方法的识别性能优于对比方法。

关键词: 人脸识别, 特征提取, 奇异值扰动, 核主成分分析, 小波变换, 最近邻分类器

Abstract:

In view of the poor performance of face recognition,an improved face recognition method based on singularvalueperturbed is proposed in this paper. Firstly, the singularvalueperturbed is applied to the single image so as to obtain expanded image set. Secondly, the wavelet decomposition is used as the preprocessing method, the lowfrequency face image is chosen as a subimage,and the highorder features are extracted by kernel principal component analysis. Finally, the nearest neighbor classifier is used for identification. The experiment results on ORL and Yale face databases show that the proposed method improves the recognition performance in comparison with the comparative approach.

Key words: face recognition;feature extract;singularvalueperturbed;kernel principal component analysis;wavelet transform;nearest neighbor classifier