An Optimal Set of Discriminant Vectors

Author

Foley, Donald H. ; Sammon, John W., Jr.

Author_Institution

Pattern Analysis and Recognition Corporation

Issue

3

fYear

1975

fDate

3/1/1975 12:00:00 AM

Firstpage

281

Lastpage

289

Abstract

A new method for the extraction of features in a two-class pattern recognition problem is derived. The main advantage is that the method for selecting features is based entirely upon discrimination or separability as opposed to the more common approach of fitting. The classical example of fitting is the use of the eigenvectors of the lumped covariance matrix corresponding to the largest eigenvalues. In an analogous manner, the new technique selects discriminant vectors (or features) corresponding to the largest "discrim-values." The new method is compared to some of the more popular alternative techniques via both data-dependent and mathematical examples. In addition, a recursive method for obtaining the discriminant vectors is given.

Keywords

Dimensionality reduction, discriminants, eigenvectors, feature extraction, feature ranking, feature selection, Karhunen-Loeve expansions, multivariate data-analysis, pattern classification, pattern recognition.; Covariance matrix; Eigenvalues and eigenfunctions; Feature extraction; Karhunen-Loeve transforms; Logic design; Pattern analysis; Pattern classification; Pattern recognition; Piecewise linear techniques; Vectors; Dimensionality reduction, discriminants, eigenvectors, feature extraction, feature ranking, feature selection, Karhunen-Loeve expansions, multivariate data-analysis, pattern classification, pattern recognition.;

fLanguage

English

Journal_Title

Computers, IEEE Transactions on

Publisher

ieee

ISSN

0018-9340

Type

jour

DOI

10.1109/T-C.1975.224208

Filename

1672801