Optimizing Feature Selection for Recognizing Handwritten Arabic Characters

Recognition of characters greatly depends upon the features used. Several features of the handwritten Arabic characters are selected and discussed. An off-line recognition system based on the selected features was built. The system was trained and tested with realistic samples of handwritten Arabic characters. Evaluation of the importance and accuracy of the selected features is made. The recognition based on the selected features give average accuracies of 88% and 70% for the numbers and letters, respectively. Further improvements are achieved by using feature weights based on insights gained from the accuracies of individual features.

Authors:

• Mohammed Z. Khedher
• Gheith A. Abandah
• Ahmed M. Al-Khawaldeh

Keywords:

• Arabic handwritten characters
• Feature extraction
• Off-line recognition
• Optical character recognition.

References:

[1] G. Abandah and F. Khundakjie, "Issues concerning code system for
Arabic letters," Dirasat Engineering Sciences J., vol. 31, no. 1, pp 165-
177, April 2004.
[2] A. Amin, "Arabic character recognition," Handbook of Character
Recognition and Document Image Analysis, pp 397-420, World
Scientific, 1997.
[3] Y. Al-Ohali, "Development and evaluation environment for typewritten
Arabic character recognition," M.Sc. thesis, King Saud University, Saudi
Arabia, 1995.
[4] Sakhr Software OCR, http://www.sakhr.com.
[5] G. Abandah and M. Khedher, "Printed and handwritten Arabic optical
character recognition - initial study," A report on research supported by
The Higher Council of Science and Technology, Jordan, August 2004.
Available: http://www.abandah.com/gheith.
[6] Handwritten Arabic Character Recognition, http://hacr.hayyan.net.
[7] R.Stefanelli and A.Rosenfeld, "Some parallel thinning algorithms for
digital pictures," J. of ACM, vol.18, no.2, pp. 255-264, 1971.
[8] S. Cha, Y. Shin and S. Srihari, "Approximate stroke sequence string
matching algorithm for character recognition and analysis," Pattern
Recognition and String Matching, vol. 13, 2003.
[9] A. Saleh, "A method of coding hand-written Arabic characters and its
application to context-free grammar," Pattern Recognition Letters, vol.
15, pp. 1265-1271, Dec. 1994.
[10] A. Amin, H. Al-sadouni, and S. Fischer, "Hand-printed Arabic character
recognition system using an artificial network," Pattern Recognition,
vol.29, no.4, pp. 663-675, April 1996.