On the Discriminant Vector Method of Feature Selection

Authors:
J. Kittler
Affiliations:
Department of Electronics, University of Southampton
Venue:
IEEE Transactions on Computers
Year:
1977

Citing 2
Cited 3

An Optimal Set of Discriminant Vectors

IEEE Transactions on Computers
Application of the Karhunen-Loève Expansion to Feature Selection and Ordering

IEEE Transactions on Computers

An extreme case of the generalized optimal discriminant transformation and its application to face recognition

Neurocomputing
Recent Developments in Pattern Recognition

IEEE Transactions on Computers
On dimensionality reduction for client specific discriminant analysis with application to face verification

SINOBIOMETRICS'04 Proceedings of the 5th Chinese conference on Advances in Biometric Person Authentication

Quantified Score

Hi-index	14.98

Visualization

Abstract

The correspondence discusses the relationship of the discriminant vector method of feature selection [1] and the method of Kittler and Young [5]. Although both methods determine the feature space coordinate axes by maximizing the generalized Fisher criterion of discriminatory power, with the exception of two class case the resulting feature spaces are considerably different because of the difference in the constraints imposed on the axes by individual methods. It is shown that the latter method is, from the point of view of dimensionality reduction, more powerful and also computationally more efficient.