首页 | 本学科首页   官方微博 | 高级检索  
     

基于聚类算法和层次支持向量机的人脸识别方法
引用本文:陈荣元,蒋加伏.基于聚类算法和层次支持向量机的人脸识别方法[J].计算技术与自动化,2006,25(1):103-105.
作者姓名:陈荣元  蒋加伏
作者单位:1. 湖南商学院现代教育技术中心,湖南,长沙,410205
2. 长沙理工大学计算机与通信工程学院,湖南,长沙,410076
基金项目:湖南省自然科学基金;湖南省教育厅科研项目
摘    要:提出一种基于聚类算法和层次支持向量机的人脸识别方法。该方法在训练阶段把利用小波变换提取的图像特征先进行聚类,再利用聚类的结果来训练层挺支持向量机;识别阶段先利用层次支持向量机初步判断待识别图像最可能所属的人,最后利用免疫算法对待识别的人脸图像进行确认。实验表明,该算法效果较佳。

关 键 词:聚类算法  层次支持向量机  免疫算法  小波变换
文章编号:1003-6199(2006)01-0103-03
收稿时间:2005-07-25
修稿时间:2005年7月25日

A Method of Face Recognition Based on Clustering Algorithm and Hierarchical Support Vector Machines
CHEN Rong-yuan,JIANG Jia-fu.A Method of Face Recognition Based on Clustering Algorithm and Hierarchical Support Vector Machines[J].Computing Technology and Automation,2006,25(1):103-105.
Authors:CHEN Rong-yuan  JIANG Jia-fu
Affiliation:1. Modern Educational Technology Center, Hunan Business College, Changsha 410205, China; 2. Institute of Computer and Communication Engineering,Changsha University of Science and Technology,Changsha 410076,China
Abstract:In this paper a method of face recognition based on clustering algorithm and hierarchical support vector machines is presented. It firstly clusters the feature vector of face image extracted using wavelet transformation then trains the hierarchical support vector machines using the clustered feature vector during the stage of the training; It firstly judges which person the image mostly belongs to using hierarchical support vector machines, the image is confirmed using immune algorithm finally during the stage of the recognition. The experiment shows that the effect is better.
Keywords:clustering algorithm  hierarchical support vector machines  immune algorithm  wavelet transformation
本文献已被 CNKI 维普 万方数据 等数据库收录!
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