DocumentCode :
2973638
Title :
Evaluating NN and HMM classifiers for handwritten word recognition
Author :
De Oliveira, JosÉ J. ; De Carvalho, JoÃo M. ; Freitas, C.O.D.A. ; Sabourin, Robert
Author_Institution :
Dept. of Electr. Eng., Fed. Univ. of Campina Grande, Brazil
fYear :
2002
fDate :
2002
Firstpage :
210
Lastpage :
217
Abstract :
This paper evaluates NN and HMM classifiers applied to the handwritten word recognition problem. The goal is analyse the individual and combined performance of these classifiers. They are evaluated considering two different combination strategies and the experiments are performed with the same database and similar feature sets. The strategy proposed takes advantage of the different but complementary mechanisms of NN and HMM to obtain a more efficient hybrid classifier. The recognition rates obtained vary from 75.9% using the HMM classifier alone to 90.4% considering the NN and HMM combination.
Keywords :
handwritten character recognition; hidden Markov models; image classification; multilayer perceptrons; optical character recognition; performance evaluation; visual databases; HMM classifiers; experiments; feature sets; handwritten word recognition; hidden markov models; image database; multilayer perceptron; neural classifiers; neural network; performance; Character recognition; Concurrent computing; Handwriting recognition; Hardware; Hidden Markov models; Neural networks; Performance analysis; Performance evaluation; Spatial databases; Writing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Graphics and Image Processing, 2002. Proceedings. XV Brazilian Symposium on
ISSN :
1530-1834
Print_ISBN :
0-7695-1846-X
Type :
conf
DOI :
10.1109/SIBGRA.2002.1167145
Filename :
1167145
Link To Document :
بازگشت