Recognition of Printed Assyrian Character Based on Neocognitron Artificial Neural Network

Nazar Saaid Sarhan¹ and Laheeb Al-Zobaidy²

¹Department of Software Engineering, Al-Isra Private University, Jordan

²Department of Computer Science, University of Mosul, Iraq 

Abstract: This paper presents the development of Assyrian machine printed character recognition system. Well-known neocognitron artificial neural network is chosen for its fast processing time and its good performance for pattern recognition problems. The average recognition rate of 95% has been achieved this confirms that the proposed neocognitron artificial neural network approach is suitable for the development of Assyrian machine printed character recognition system. 

Keywords: Assyrian machine printed character recognition, neocognitron artificial neural network, whitespace, feature extraction.