| 支持向量机的人脸识别方法 |
| |
| 引用本文： | 周志明,陈敏.支持向量机的人脸识别方法[J].咸宁学院学报,2003,23(3):19-22. |
| |
| 作者姓名： | 周志明 陈敏 |
| |
| 作者单位： | 1. 华中科技大学,数学系,湖北,武汉,430074
2. 咸宁学院,数学系,湖北,咸宁,437005 |
| |
| 摘 要： | 简要介绍了基于统计学习理论中结构风险最小化原则的支撑向量机(SVMs)技术的国内外研究现状,分析了该技术的优越性和存在的某些局限,并提出了该技术的一些改进思路.
|
| 关 键 词： | 支持向量机 结构风险 分类超平面 核函数 |
| 文章编号： | 1006-5342(2003)03-0019-04 |
| 修稿时间： | 2003年4月22日 |
A Face Recognition Method Using Support Vector Machines |
| |
| ZHOU Zhi-ming ,CHEN Min.A Face Recognition Method Using Support Vector Machines[J].Journal of Xianning College,2003,23(3):19-22. |
| |
| Authors: | ZHOU Zhi-ming CHEN Min |
| |
| Institution: | ZHOU Zhi-ming 1,CHEN Min 2 |
| |
| Abstract: | The support vector machines (SVMs) based on structural risk minimization (SRM) principle in the statistical learning theory is an effective method to solve the small sample classify problem and has been the study focus in pattern recognition. Its main ideas is using the best classify hyper plane selected in a hyper planes set as the classification function. If the sample is nonlinear, map it into a higher dimension space (call "feature space") via a nonlinear mapping function and thus can execute the nonlinear classify as in the input space. In this paper, we introduce the study focuses for the SVMs civil and abroad, discuss the advantages and disadvantages of the SVMs and propose some study ideas in the future work. |
| |
| Keywords: | Support vector machines Structural risk Classify hyper plane Kernel function |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
