Optimizing the recognition rates of unconstrained handwritten numerals using biorthogonal spline wavelets

Author

Correia, Suzete E N ; De Carvalho, Joao M.

Author_Institution

Dept. de Engenharia Eletrica, Univ. Federal da Paraiba, Joao Pessoa, Brazil

Volume

2

fYear

2000

fDate

2000

Firstpage

251

Abstract

In this paper an approach for off-line recognition of unconstrained handwritten numerals is presented. This approach uses the Cohen-Daubechies-Feauveau (CDF) family of biorthogonal spline wavelets as a feature extractor for absorbing local variations in handwritten characters and a multilayer cluster neural network as classifier. Experiments with the bases CDF 2/2, CDF 2/4, CDF 3/3 and CDF 3/7 were performed using the handwritten numeral database of Concordia University of Canada. The results show that CDF biorthogonal wavelets yield a performance improvement of 2.4% in numeral recognition, compared to the results obtained with the Haar wavelets

Keywords

feature extraction; feedforward neural nets; handwritten character recognition; pattern classification; splines (mathematics); wavelet transforms; Cohen-Daubechies-Feauveau family; Concordia University of Canada; biorthogonal spline wavelets; cluster neural network; feature extraction; handwritten character recognition; handwritten numerals; multilayer neural network; pattern classification; Feature extraction; Filters; Frequency; Handwriting recognition; Multi-layer neural network; Neural networks; Spatial databases; Spline; Wavelet analysis; Wavelet transforms;

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.2000.906060

Filename

906060