Offline Arabic Text Recognition System

Optical character recognition (OCR) systems provide human-machine interaction and are widely used in many applications. Much research has already been done on the recognition of Latin, Chinese and Japanese characters. Against this background, it has been experienced that only few papers have specifically addressed to the problem of Arabic text recognition and languages using Arabic script like Urdu and Parsi. This is due to the lack of interest in this field and in part due to the complex nature of the Arabic language. This paper presents a technique for the automatic recognition of Arabic printed text using artificial neural networks. The main features of the system are preprocessing of the text, segmentation of the text to individual characters, feature extraction using moment invariant technique and recognition using RBF network.