Title :
A minimum distance feature effectiveness criterion (Corresp.)
Author :
Fukunaga, Kaori ; Krile, T.
fDate :
9/1/1968 12:00:00 AM
Abstract :
A new criterion for measuring separability of two pattern classes is given that is based on the determination of an averaged minimum distance from the mean feature vectors of two classes to an optimal separating hypersurface determined by Baye´s postulate.
Keywords :
Feature extraction; Pattern classification; Aging; Bismuth; Covariance matrix; Density measurement; H infinity control; Pattern classification; Probability density function; Research and development; Scattering;
Journal_Title :
Information Theory, IEEE Transactions on
DOI :
10.1109/TIT.1968.1054204
