首页 | 本学科首页   官方微博 | 高级检索  
     

一种具有统计不相关性的最佳鉴别矢量集
引用本文:金忠,杨静宇,陆建峰.一种具有统计不相关性的最佳鉴别矢量集[J].计算机学报,1999,22(10):1105-1108.
作者姓名:金忠  杨静宇  陆建峰
作者单位:南京理工大学计算机科学系,南京,210094
摘    要:在模式识别领域,基于Fisher鉴别准则语数的Foley-Sammon最佳鉴别技术在着重大的影响,特征抽取的一般原则是最好抽取模式的不相关的特征,而Foley-Sammon最佳鉴别矢量集的诸鉴别特是统计相关的。

关 键 词:模式识别  特征抽取  鉴别分析  人脸识别
修稿时间:1998年11月3日

AN OPTIMAL SET OF UNCORRELATED DISCRIMINANT FEATURES
JIN Zhong,YANG Jing-Yu,LU Jian-Feng.AN OPTIMAL SET OF UNCORRELATED DISCRIMINANT FEATURES[J].Chinese Journal of Computers,1999,22(10):1105-1108.
Authors:JIN Zhong  YANG Jing-Yu  LU Jian-Feng
Abstract:Based on Fisher's discriminant function, Foley Sammon's optimal set of discriminant vectors has great influence in the area of pattern recognition. A general rule for feature extraction is to extract features as uncorrelated as possible, but the discriminant vectors in Foley Sammon's optimal set are correlated. In the paper here, an optimal set of uncorrelated discriminant features is presented, and some formula to compute it is given. Experiments with ORL face image database have been performed. Experimental results show that new optimal set of uncorrelated discriminant features has powerful ability of feature extraction, and is superior to Foley Sammon's optimal set of discriminant vectors.
Keywords:Pattern recognition  feature extraction  discriminant analysis  face recognition  
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