Title

Some approaches to pattern recognition problems.

Authors

A. Golodnikov; P. Knopov; V. Pepelyaev

Abstract

Abstract  A new approach to selecting the Gibbs distribution in models of objects to be recognized is proposed. This approach proposes to determine the lower and upper bounds for probabilities of the object under study. The distance between these bounds may be used as a measure of error in pattern recognition problems.

Subjects

PATTERN perception; STATISTICAL correlation; PROBABILITY theory; GRAPHIC methods in statistics

Publication

Cybernetics & Systems Analysis, 2007, Vol 43, Issue 6, p810

ISSN

1060-0396

Publication type

Academic Journal

DOI

10.1007/s10559-007-0105-z