Isolated Handwritten Eastern Arabic Numerals Recognition Using Support Vectors Machines

B. El Kessab (Faculty of Science and Technology, BP 523, BeniMellal)
C. Daoui (Faculty of Science and Technology, BP 523, BeniMellal)
B. Bouikhalene (Faculty of Science and Technology, BP 523, BeniMellal)
R. Salouan (Faculty of Science and Technology, BP 523, BeniMellal)

Indonesian Journal of Electrical Engineering and Computer Science
Vol 15, No 2: August 2015

In this paper, we present a comparison between the different variations of virtual retina (grid size) in features extraction with the support vectors machines classifier for isolated handwritten Eastern Arabic numerals recognition. For this purpose we have used for pre-processing each numeral image the median filter, the thresholding, normalization and the centering techniques. Furthermore, the experements results that we have obtained demonstrate really that the most powerful method is that virtual retina size equal 20x20. This work has achieved approximately 85% of success rate for Eastern Arabic numerals database identification.