A Methodology for Handwritten Character Recognition Using SVM

Abstract

This paper discusses a methodology for handwritten character recognition applying feature subset selection to reduce number of features. Its novelty lies in the use of a genetic algorithm for the preparation of input data for a support vector machine which is employed to recognize the handwritten Persian digits in particular. Comprehensive experiments on handwritten Persian digits demonstrate the capability of such a classifier running on reduced extracted features.

Extracted Key Phrases

1 Figure or Table

Showing 1-10 of 17 references

Applying OWA Operator in a Multi-Agent Architecture of an Emotional Robot

  • M Shayganfar, B Moshiri, C Lucas
  • 2006

An Introduction to Support Vector Machines

  • N Cristianini, J Shawe-Taylor
  • 2000