A learning method for definite canonicalization based on minimum classification error

Author

SATo, Atsushi

Author_Institution

Comput. & Commun. Media Res., NEC Corp., Kawasaki, Japan

Volume

2

fYear

2000

fDate

2000

Firstpage

199

Abstract

This paper presents a novel learning method for definite canonicalization (DC) based on minimum classification error (MCE). It is shown that DC is identical to normalized cross-correlation, and that the complementary similarity measure is derived from DC for binary patterns. The proposed learning method is derived from the framework of generalized learning vector quantization (GLVQ), which is one of the discriminative learning methods based on MCE. Experimental results obtained for machine-printed Kanji character recognition reveal that the proposed method achieves high performance recognition of low-quality patterns

Keywords

correlation methods; learning (artificial intelligence); minimisation; optical character recognition; pattern classification; vector quantisation; DC; GLVQ; LVQ; MCE; VQ; complementary similarity measure; definite canonicalization; discriminative learning methods; generalized learning vector quantization; high performance recognition; low-quality patterns; machine-printed Kanji character recognition; minimum classification error; normalized cross-correlation; Character recognition; Computer errors; Degradation; Electronic mail; Learning systems; Marine vehicles; National electric code; Pattern recognition; Robustness; Vector quantization;

fLanguage

English

Publisher

ieee

Conference_Titel

Pattern Recognition, 2000. Proceedings. 15th International Conference on

Conference_Location

Barcelona

ISSN

1051-4651

Print_ISBN

0-7695-0750-6

Type

conf

DOI

10.1109/ICPR.2000.906047

Filename

906047