Title :
Recognition of unconstrained handwritten numerals using biorthogonal spline wavelets
Author :
Correia, Suzete E N ; De Carvalho, JoÃo M.
Author_Institution :
Dept. de Engenharia Eletrica, Univ. Fed. da Paraiba, Joao Pessoa, Brazil
Abstract :
The work proposes a novel approach for recognition of unconstrained handwritten numerals using the biorthogonal spline wavelets Cohen-Daubechies-Feauveau (CDF) 3/7 as a feature extractor and a multilayer cluster neural network as a classifier. Experiments with the CENPARMI database show that this method yields good results
Keywords :
feature extraction; handwritten character recognition; image classification; neural nets; splines (mathematics); wavelet transforms; CENPARMI database; biorthogonal spline wavelets; classifier; multilayer cluster neural network; unconstrained handwritten numeral recognition; Character recognition; Data mining; Feature extraction; Handwriting recognition; Multi-layer neural network; Neural networks; Shape; Spatial databases; Spline; Wavelet transforms;
Conference_Titel :
Computer Graphics and Image Processing, 2000. Proceedings XIII Brazilian Symposium on
Conference_Location :
Gramado
Print_ISBN :
0-7695-0878-2
DOI :
10.1109/SIBGRA.2000.895826
