Différenciation d'écritures arabe et latine de natures imprimée et manuscrite par approche globale
Résumé
In this paper, we treat the problem of script differentiation. Our aim is to use the global approach to resolve the problem of differentiation between the Arabic and the Latin scripts in printed and handwritten natures. In our method, features are extracted using three tools: Gabor filters, gray-level co-occurrence matrices and wavelets. The results of differentiation obtained on a base of 800 Arab and Latin documents, printed and handwritten are presented. The highest rate of correct identification is obtained with gray-level cooccurrence matrices (84.75%). We can consider this rate as satisfactory taking into account the simplicity of the approach and the complexity of the treated problem.
Origine : Fichiers produits par l'(les) auteur(s)
Loading...