Handwritten recognition with multiple classifiers for restricted lexicon

Author

de Oliveira, J.J., Jr. ; Kapp, M.N. ; Freitas, C.O.A. ; de Carvalho, J.M. ; Sabourin, R.

Author_Institution

Coordenacao de Pos-Graduacao em Engenharia Eletrica, Univ. Fed. de Campina Grande, Brazil

fYear

2004

fDate

17-20 Oct. 2004

Firstpage

82

Lastpage

89

Abstract

This paper presents a multiple classifier system applied to the handwritten word recognition (HWR) problem. The goal is to analyse the influence of different global classifiers taken in isolation as well as combined in a particular HWR task. The application proposed is the recognition of the Portuguese handwritten names of the months. The strategy takes advantage of the complementary mechanisms of three different classifiers: conventional neural network, class-modular neural network and hidden Markov models, yielding a multiple classifier that is more efficient than either individual technique. The recognition rates obtained vary from 75.9% using the standalone HMM classifier to 96.0% considering the classifier combination.

Keywords

handwritten character recognition; hidden Markov models; neural nets; pattern classification; Portuguese handwritten names; class-modular neural network; conventional neural network; handwritten word recognition; hidden Markov models; month recognition; multiple classifier system; multiple classifiers; restricted lexicon; Computer graphics; Feature extraction; Handwriting recognition; Hidden Markov models; Humans; Image databases; Isolation technology; Neural networks; Speech recognition; System performance;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Graphics and Image Processing, 2004. Proceedings. 17th Brazilian Symposium on

ISSN

1530-1834

Print_ISBN

0-7695-2227-0

Type

conf

DOI

10.1109/SIBGRA.2004.1352947

Filename

1352947