Authors: Cveta Martinovska; Mimoza Klekovska; Igor Nedelkovski; Dragan Kaevski
Addresses: Computer Science Faculty, University Goce Delcev, Tosho Arsov 14, 2000 Stip, Macedonia ' Faculty of Technical Sciences, University St. Kliment Ohridski, Ivo Ribar Lola bb, 7000 Bitola, Macedonia ' Faculty of Technical Sciences, University St. Kliment Ohridski, Ivo Ribar Lola bb, 7000 Bitola, Macedonia ' Faculty of Electrical Engineering and Information Technology, University St. Cyril and Methodius, Rugjer Boshkovik bb, 1000 Skopje, Macedonia
Abstract: This paper describes two novel methodologies for recognition of old Slavic Cyrillic characters. The first recognition method is based on a decision tree classifier and the second one uses a fuzzy classifier. Both methods use the same set of features extracted from the character bitmaps. The prototypes are obtained by applying the logical operators on the samples of digitalised characters from original manuscripts. According to the experimental results relevant features for defining a particular character are number and position of spots in the outer segments, presence and position of horizontal and vertical lines and holes, compactness and symmetry. The fuzzy classifier creates a prototype which consists of fuzzy rules by means of fuzzy aggregation of character features. The classifier based on a decision tree is realised by a set of rules. We have implemented the proposed classifiers and have experimentally tested their efficiency calculating their recognition accuracy and precision.
Keywords: classifiers; decision tree; fuzzy logic; recognition models; old Slavic Cyrillic alphabet; computational intelligence; character recognition; Cyrillic characters; feature extraction; character bitmaps.
International Journal of Computational Intelligence Studies, 2013 Vol.2 No.3/4, pp.264 - 287
Available online: 17 Nov 2013 *Full-text access for editors Access for subscribers Purchase this article Comment on this article