• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学 ›› 2010, Vol. 32 ›› Issue (5): 34-36.

• • 上一篇    下一篇

正交约束的无监督统计不相关最佳鉴别平面

曹苏群1,2,王骏1,3,王士同1   

  1. (1.江南大学信息学院,江苏 无锡 214122;2.淮阴工学院机械工程学院,江苏 淮安 223003; 3.南京理工大学计算机科学与技术学院,江苏 南京 210093)
  • 收稿日期:2009-09-15 修回日期:2009-12-08 出版日期:2010-04-28 发布日期:2010-05-11
  • 通讯作者: 曹苏群 E-mail:caosuqun@126.com
  • 作者简介:曹苏群(1976),男,江苏淮安人,博士生,研究方向为模式识别和机器学习;王骏,博士,研究方向为模式识别、数字图像处理和智能信息处理;王士同,教授,博士生导师,研究方向为人工智能、机器学习和生物信息学。
  • 基金资助:

    国家863计划资助项目(2007AA1Z158); 国家自然科学基金重点项目(60773206,60704047); 江苏省高校自然科学重大基础研究资助项目(09KJA460001);江苏省高校自然科学重大基础研究资助项目(09KJA460001);淮安市国际科技合作项目(HG004);淮阴工学院青年科技基金资助项目(HGQN0701)

The Unsupervised and Uncorrelated Optimal Discriminant Plane Based on Orthogonal Constraints

CAO Suqun1,2,WANG Jun1,3 ,WANG Shitong1   

  1. (1.School of Information,Jiangnan University,Wuxi 214122;
    2.Faculty of Mechanical Engineering,Huaiyin Institute of Technology,Huaian 223003;
    3.School of Computer Science and Technology,Nanjing University of Science and Technology,Nanjing 210093,China)
  • Received:2009-09-15 Revised:2009-12-08 Online:2010-04-28 Published:2010-05-11
  • Contact: CAO Suqun E-mail:caosuqun@126.com

摘要:

赵海涛等提出的改进的最佳鉴别平面(IODP)只能用于有监督模式,基于此,本文提出将IODP扩展到无监督模式下的方法。在优化模糊Fisher准则求取第一条最佳鉴别矢量的基础上,求取同时满足正交约束与模糊总体散布矩阵共轭正交约束的第二条最佳鉴别矢量,构成正交约束的无监督统计不相关最佳鉴别平面(OUUODP),进而获得一种新的无监督特征抽取方法。对CMUPIE人脸数据库进行实验,结果表明,当类别差异较大时,该方法能够抽取有利于分类的特征,获得了优于主成分分析与独立成分分析方法的性能。

关键词: 无监督模式, 特征降维, 最佳鉴别平面, 人脸识别

Abstract:

An improved optimal discriminant plane(IODP) proposed by Zhao can only be used in the supervised pattern. Based on this point, a novel method is presented to extend IODP to the unsupervised pattern. On the basis of optimizing the fuzzy Fisher criterion to calculate the first optimal discriminant vector, the second optimal discriminant vector with the orthogonal constraint and the conjugated orthogonal constraint of the fuzzy totalclass scatter matrix can be figured out. These two vectors constitute the orthogonalconstraintbased unsupervised and uncorrelated optimal discriminant plane(OUUODP). With these, a novel unsupervised feature extraction method is obtained. The experimental results for the CMUPIE face database demonstrate that this method can extract the features which are conducive to classification and is superior to principal component analysis and independent component analysis when the betweenclass difference is big.

Key words: unsupervised pattern, feature dimension reduction, optimal discriminant plane, face recognition

中图分类号: