Saudi Banknote Recognition Using Inductive Learning

Abstract

This paper presents an alternative technique for Saudi Banknote recognition. The technique employs template matching for feature extraction and RULES3 inductive learning algorithm to extract the necessary set of rules and to recognize a banknote. 10 of 3x3 masks were used to represent a banknote. Each banknote (or example) is presented by the frequencies of the masks used. The system was trained for back and front sides of 8 different Saudi banknotes. Ten images were used for each side of the banknotes for learning purpose. The system has been tested using many “unseen” examples and correctly classified all of them.