Character recognition has been widely understood as a means of mechanizing the process of understanding text in the written form to facilitate fast and efficient use of text. Indeed, text existing all around us presents information for peoples. However, tourists in foreign countries are unable to understand what indicate text on road signs, shop names, product advertisements, posters, etc. when they are unfamiliar with the native language of the visited country.
Currently there is no available dataset of Arabic script text images in the wild. Since our aim is to help the research community in standardizing the evaluation of scene Arabic text recognition, the Tunisian Research Groups in Intelligent Machines of University of Sfax (REGIM lab of Sfax) will provide the Arabic Scene Text Image Database of Regim Lab (ARASTI database) freely of charge to mainly Arabic scene character recognition researchers and to increase total of researches done to enhance scene character recognition.
All documents and papers that uses the Arabic Scene Text Image Database of Regim Lab (ARASTI database) will acknowledge the use of the database by including an appropriate citation to the following:
: Maroua Tounsi, Ikram Moalla, Adel M. Alimi, "ARASTI: A database for Arabic scene text recognition." 2017 1st International Workshop on Arabic Script Analysis and Recognition (ASAR). IEEE, 2017.
: Maroua Tounsi, Ikram Moalla, Adel M. Alimi, Frank Lebouregois, “Feature Representation using Sparse Coding for Robust Arabic Characters Recognition in Natural Scenes” , ICDAR 2015.
Download Zip file and extract it.