Les Supports Vecteurs Machines (SVM) pour la reconnaissance des caractères manuscrits arabes

Abstract

Optic Character Recognition (OCR) has a main role in the present time. It's capable to resolve complex problems and simplify human activities. The OCR yields to 70's, since many solutions had proposed, and they are varied as Latin methods. This work presents a survey concerning Arabic Optic Character Recognition (AOCR). A general study of handwritten character recognition systems is developed, then it takes apart the segmentation phase because it is a crucial phase of the recognition process. We had presented a state of the art of the character segmentation methods, after that a view of the OCR area, Arabic language, and some of Arabic writing normalization problems are presented. After a comparison between Arabic handwritten characters segmentation methods, we had proposed a contribution through a segmentation algorithm.