On Optimal Pairwise Linear Classifiers for Normal Distributions: The Two-Dimensional Case.

Luís G. Rueda B. John Oommen

Published in: IEEE Trans. Pattern Anal. Mach. Intell. (2002)

Keyphrases

linear classifiers
normal distribution
pairwise
multi class
closed form
hyperplane
data sets
optimal solution
higher order
markov random field
random variables
computer vision
generalization error
standard deviation
covariance matrix
similarity measure
sample size
high dimensional data
worst case
pattern recognition