Recognition of printed Arabic words with fuzzy ARTMAP neural network

This paper presents a new method for the recognition of Arabic text using global fmtures and Fuzzy ARTMAP neural network The method is divided into three major steps. The first step is digitization and pre-processing to create connected component. The secondstep is concerned with fmture extraction, where global f w r e s of the input word are used to extract features such as number of suhvorch, number of peaks within the subword, number andposition of the complementary character, etc., to avoid the dij?cul@ of segmentation stage. The third step is the classification and is composed of a single Fuzzy A R T W . The method was evaluated with 3255 images of 21 7 Arabic work with dixerent fonts (each word has 15 samples), and the mean correct classification rate was 95.25%.