Neural versus syntactic recognition of handwritten numerals

Author

Veloso, Luciana R. ; De Carvalho, João Marques

Author_Institution

Dept. de Engenharia Electrica, UFPB, Campina Grande, Brazil

fYear

1999

fDate

20-22 Sep 1999

Firstpage

233

Lastpage

236

Abstract

This work concerns the analysis, implementation and evaluation of three different methods for handwritten numerical character recognition. The first approach uses a classifier based on syntactical analysis by decision tree. The other two methods consist of: (a) a conventional feedforward multilayer neural network; and (b) a recurrent neural network, for which the elements of the output layer are all interconnected. The CENPARMI database was utilized for evaluation of the systems

Keywords

computational linguistics; decision trees; feedforward neural nets; handwritten character recognition; image recognition; recurrent neural nets; CENPARMI database; conventional feedforward multilayer neural network; decision tree; handwritten numerals; handwritten numerical character recognition; neural recognition; output layer; recurrent neural network; syntactic recognition; syntactical analysis; Character recognition; Feedforward neural networks; Handwriting recognition; Machine intelligence; Multi-layer neural network; Neural networks; Postal services; Shape; Spatial databases; System testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Document Analysis and Recognition, 1999. ICDAR '99. Proceedings of the Fifth International Conference on

Conference_Location

Bangalore

Print_ISBN

0-7695-0318-7

Type

conf

DOI

10.1109/ICDAR.1999.791767

Filename

791767