A New Recognition Scheme for Machine- Printed Arabic Texts based on Neural Networks

Authors: Z. Shaaban

Abstract:

This paper presents a new approach to tackle the problem of recognizing machine-printed Arabic texts. Because of the difficulty of recognizing cursive Arabic words, the text has to be normalized and segmented to be ready for the recognition stage. The new scheme for recognizing Arabic characters depends on multiple parallel neural networks classifier. The classifier has two phases. The first phase categories the input character into one of eight groups. The second phase classifies the character into one of the Arabic character classes in the group. The system achieved high recognition rate.

Keywords: Neural Networks, character recognition, feature extraction, multiple networks, Arabic text.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1331183

References:

[1] I.S. Abuhaiba and P. Ahmed, "Restoration of Temporal Information in Off-Line Arabic Handwriting," Pattern Recognition(26), No. 7, July 1993, pp. 1009-1017.
[2] M. Altuwaijri and A. Bayoumi, "A new recognition system for multifont Arabic Cursive words, " Proceedings of ICECS'95, Amman- Jordan, pp.298-303,1995.
[3] A. Amin, H.B. Al-Sadoun and S. Fischer, "Hand-Printed Arabic Character-Recognition System Using an Artificial Network," Pattern Recognition(29), No. 4, April 1996, pp. 663-675.
[4] J. Cao, M. Ahmadi and M. Shridhar, "Recognition of handwritten numerals with multiple feature and multistage classifier," Pattern Recognition,1995, vol.28, no.2, pp. 153-160.
[5] S. G├╝nter and H. Bunke, "Multiple classifier systems in off-line handwritten word recognition - on the influence of training set and vocabulary size," Int. Journal of Pattern Recognition and Art. Intelligence, 2004, vol. 18, no. 7, pages 1303 - 1320.
[6] F. Kimura and M. Shridhar, "Handwritten numerical recognition based on multiple algorithms," Pattern Recognition,1991, vol.24, no.lO.pp.969-983.
[7] Z. Shaaban and G. Sulong, "Uppercase hand-printed character recognition using parallel neural architecture," Proc. of the 1ASTED International conference modeling and simulation Pittsburgh-USA, 1995 pp.307-309
[8] Z. Shaaban and Z. Sulong, "Recognition of connected handwritten characters based on moments invariants using neural networks," Proceedings of ACCV'95 Second Asian Conference on Computer Vision, 1995, Singapore pp.l- 335-339
[9] Z. Shaaban, G. Sulong and B. Duin, "Recognition of handprinted characters using distance transform and moment invariants via parallel neural networks," Proceedings of ICECS'95 International conference on electronics, circuits and systems '95, 1995, Amman-Jordan, pp.393- 398.
[10] Z. Shaaban, G. Sulong and B. Duin, "Symbol recognition based on distance transform," Proc. of the IASTED International conference on signal and image processing and applications, 1996 Annecy-France, pp.225-230.
[11] S. N. Srihari, "Recognition of handwritten and machine printed text of postal address interpretation," In (4).
[12] C. Y. Suen, "Computer recognition of unconstrained handwritten numerals," Proceedings of the IEEE vol.80, no.7, pp.1162-1180,1992.
[13] L. Xu, A. Krzyzak and C. Y. Suen, "Methods of combining multiple classifiers and their applications to handwriting recognition," IEEE transactions on systems, man, and Cybernetics vol.22, no.3, pp.418- 435,1992.
[14] Z. Shaaban, Algorithms for off-line upper-case hand-written text recognition and its associated processes, Ph.D. Thesis, University of Technology Malaysia, 1996.