计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (25): 154-156.DOI: 10.3778/j.issn.1002-8331.2009.25.047

• 图形、图像、模式识别 • 上一篇    下一篇

结合独立成分分析和核向量机进行人脸识别

彭中亚,程国建,曹庆年   

  1. 西安石油大学 计算机学院,西安 710065
  • 收稿日期:2009-04-28 修回日期:2009-06-24 出版日期:2009-09-01 发布日期:2009-09-01
  • 通讯作者: 彭中亚

Face recognition by combining independent component analysis with core vector machines

PENG Zhong-ya,CHENG Guo-jian,CAO Qing-nian   

  1. School of Computer Science,Xi’an Shiyou University,Xi’an 710065,China
  • Received:2009-04-28 Revised:2009-06-24 Online:2009-09-01 Published:2009-09-01
  • Contact: PENG Zhong-ya

摘要: 在人脸识别过程中,首先利用独立成分分析得到独立的人脸基影像,所提取的特征就是人脸图像在基影像上的投影系数,通过选择合适的特征个数可以达到较高的识别准确率。然后采用支持向量机和核向量机分别对待识别图像在基影像上的投影系数进行分类判决,结果显示二者都能达到较高的识别准确率,但随着特征个数的增加,核向量机的准确率更高,训练时间更短,支持向量更少。实验表明方法可行有效的。

关键词: 人脸识别, 独立成分分析, 核向量机, 支持向量机

Abstract: In the process of facing recognition,Independent Component Analysis(ICA) is used to extract face feature,which is the coefficient of face images projecting to the independent base images.Faces can be recognized by classifying the coefficients using Support Vector Machines(SVM) and Core Vector Machines(CVM).Both SVM and CVM have high recognition accuracy.But with the increase of ICA feature number,CVM has higher accuracy,less training time and fewer support vectors.Experimental results show that the algorithm is feasible and effective for face recognition.

Key words: face recognition, independent component analysis, core vector machines, support vector machines

中图分类号: