On the performances of neuronal classifiers for pattern recognition

Author

Boughrara, Hayet ; Chtourou, Mohamed

Author_Institution

Sfax Eng. Sch., Univ. of Sfax, Safagis

fYear

2009

fDate

23-26 March 2009

Firstpage

1

Lastpage

4

Abstract

This work consists on the evaluation of the performances of three neural classifiers. The Multi-Layer Perceptron (MLP), the Self-Organizing Map (SOM), the Learning Vector Quantization (LV Q) are considered by this study. The example that will be considered in the evaluation of the technical classifications´s performances is the handwritten character recognition.

Keywords

geometry; handwritten character recognition; learning (artificial intelligence); multilayer perceptrons; pattern classification; self-organising feature maps; geometric moment invariant; handwritten character recognition; learning vector quantization; multilayer perceptron; neuronal classifier; pattern recognition; self-organizing map; Character recognition; Data mining; Design engineering; Design optimization; Intelligent control; Multilayer perceptrons; Pattern recognition; Performance evaluation; Signal design; Vector quantization; Classification; Learning Vector Quantization LV Q; Multi-layer Perceptron MLP; Self-OrganizingMap SOM; Zernike moment; geometric invariant moment; handwriting character recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Systems, Signals and Devices, 2009. SSD '09. 6th International Multi-Conference on

Conference_Location

Djerba

Print_ISBN

978-1-4244-4345-1

Electronic_ISBN

978-1-4244-4346-8

Type

conf

DOI

10.1109/SSD.2009.4956752

Filename

4956752