Open AccessIsolated Handwritten Eastern Arabic Numerals Recognition Using Support Vectors Machines
Author(s) -
B. El Kessab,
C. Daoui,
B. Bouikhalene,
R. Salouan
Publication year - 2015
Publication title -
telkomnika: indonesian journal of electrical engineering/telkomnika
Language(s) - English
Resource type - Journals
eISSN - 2460-7673
pISSN - 2302-4046
DOI - 10.11591/tijee.v15i2.1548
Subject(s) - computer science , numeral system , thresholding , normalization (sociology) , artificial intelligence , pattern recognition (psychology) , arabic , support vector machine , classifier (uml) , speech recognition , arabic numerals , feature extraction , preprocessor , image (mathematics) , linguistics , philosophy , sociology , anthropology
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.