首页 | 本学科首页   官方微博 | 高级检索  
     

基于不同Margin的人脸特征选择及识别方法
引用本文:李伟红,陈伟民,杨利平,龚卫国.基于不同Margin的人脸特征选择及识别方法[J].电子与信息学报,2007,29(7):1744-1748.
作者姓名:李伟红  陈伟民  杨利平  龚卫国
作者单位:重庆大学光电技术及系统教育部重点实验室,重庆,400044
基金项目:教育部科学技术研究项目 , 重庆市自然科学基金 , 重庆市自然科学基金
摘    要:Margin在机器学习中具有很重要的意义,基于margin的特征选择方法就是从分类的角度对特征集各特征的权重进行分析。该文对不同的margin进行了分析,提出将sample-margin和hypothesis-margin分别作为特征选择标准对SBS特征选择方法进行改进,然后设计具有最佳超参数的SVM多项式分类器进行人脸识别。实验在FRERT人脸图像库上进行并与Relief特征选择方法进行了比较,对SVM和NN分类器的实验结果也进行了分析。实验结果显示:该文提出的人脸识别特征选择及识别方法是有效、适用的。

关 键 词:人脸识别  Margin  特征选择  支持向量机(SVM)  顺序后退法(SBS)
文章编号:1009-5896(2007)07-1744-05
收稿时间:2005-12-5
修稿时间:2005-12-052006-04-03

Face Feature Selection and Recognition Based on Different Types of Margin
Li Wei-hong,Chen Wei-min,Yang Li-ping,Gong Wei-guo.Face Feature Selection and Recognition Based on Different Types of Margin[J].Journal of Electronics & Information Technology,2007,29(7):1744-1748.
Authors:Li Wei-hong  Chen Wei-min  Yang Li-ping  Gong Wei-guo
Affiliation:Key Lab of Optoelectronic Technology and Systems of Education Ministry of China, Chongqing University, Chongqing 400044, China
Abstract:Margin plays an important role in research of machine learning. Margin-based feature selection methods choose the weights of features from the view of classification. This paper analyzes different types of margin and proposed methods to improve the Sequential Backward Selection (SBS) method respectively using sample-margin and hypothesis-margin as feature selection criterion. A SVM polynomial classifier, which has optimal hyper-parameters, is then designed for face recognition. Experiments are conducted on FERET face database. Recognition accuracies between the proposed methods and relief feature selection method are compared. Experiments are also conducted by respectively using SVM and Nearest Neighbor (NN) classifier. Experimental results indicate that the proposed feature selection and recognition methods are efficient for face recognition.
Keywords:Face recognition  Margin  Feature selection  Support Vector Machine (SVM)  Sequential Backward Selection (SBS)
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《电子与信息学报》浏览原始摘要信息
点击此处可从《电子与信息学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号