计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (29): 104-106.

• 产品、研发、测试 • 上一篇    下一篇

多类SVM分类方法在智能像卡识别中的实现

张 欣,戴 永

  

  1. 湘潭大学 信息工程学院,湖南 湘潭 411105
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-10-11 发布日期:2007-10-11
  • 通讯作者: 张 欣

Implementation of SVM multi-class classification in intelligent image card recognition

ZHANG Xin,DAI Yong   

  1. College of Information & Engineering,Xiangtan University,Xiangtan,Hunan 411105,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-10-11 Published:2007-10-11
  • Contact: ZHANG Xin

摘要: 支持向量机是在统计学习理论基础上发展起来的一种新的机器学习方法,在模式识别、回归分析、函数估计等领域有着广泛的应用。论文提出在单片机系统上实现这一算法的方法,并在智能像卡联网门禁系统中得以实现。应用结果表明,该方法使像卡识别在获得SVM多类分类识别能力的同时,也有效降低了单片机的计算负荷。

关键词: 支持向量机, 二叉树, 多类分类, 智能像卡, 联网门禁

Abstract: SVM(Support Vector Machine) is a new machine-learning method which is developed based on statistical theory and has many applications in pattern recognition,regression analysis,function evaluation,etc.This paper proposes a method of implementation of the algorithm in Network Entrance Guard System using IIC(Intelligent Image Card).Numerical experiments on large problems demonstrate the method not only acquires classification capability of SVM,but also reduces computing tasks for IIC recognition.

Key words: Multi-class Support Vector Machines, binary tree, multi-class classification, Intelligent Image Card, Networking Entrance Guard System