Logo LVSN
EnglishAccueil
A proposPersonnesRecherchePublicationsEvenementsProfil
A propos
Publications

 

 

 

 

CERVIM

REPARTI

MIVIM

Genetic Engineering of Hierarchical Fuzzy Regional Representations for Handwritten Character Recognition


Christian Gagné and Marc Parizeau


Abstract - This paper presents a genetic programming based approach for optimizing the feature extraction step of a handwritten character recognizer. This recognizer uses a simple multilayer perceptron as a classifier and operates on a hierarchical feature space oforientation, curvature, and center of mass primitives. The nodes of the hierarchy represent rectangular sub-regions of their parent node, the tree root corresponding to the character's bounding box. Within each sub-region, a variable number of fuzzy features are extracted. Genetic programming is used to simultaneously learn the best hierarchy and the best combination of fuzzy features. Moreover, the fuzzy features are not predetermined, they are inferred from the evolution process which runs a two-objective selection operator. The first objective maximizes the recognition rate, and the second minimizes the feature space size. Results on Unipen data show that, using this approach, robust representations could be obtained that out-performed comparable humandesigned hierarchical fuzzy regional representations.

download document

Bibtex:

@article{Gagné607,
    author    = { Christian Gagné and Marc Parizeau },
    title     = { Genetic Engineering of Hierarchical Fuzzy Regional Representations for Handwritten Character Recognition },
    volume    = { 8 },
    number    = { 4 },
    pages     = { 223-231 },
    year      = { 2006 },
    month     = { September },
    journal   = { International Journal of Document Analysis and Recognition }
}

Dernière modification: 2006/01/24 par parizeau

     
   
   

©2002-. Laboratoire de Vision et Systèmes Numériques. Tous droits réservés