Intelligent Banknotes Recognition Based on Support Vector Machine (SVM)
Shafaf Ibrahim · International Journal of Emerging Trends in Engineering Research · 2020
Banknotes recognition is one type of intelligent system that is essential in today's modern world.Yet, it remains as a challenging task as the banknote may suffer from defects and the images are distorted during acquisition, resulting the need for a robust recognition system to mitigate these shortcomings.Additionally, more than 200 different currencies are being used in different countries around the world which might lead to confusion in differentiating these banknotes visually.Thus, this paper proposed a study of intelligent banknotes recognition based on Support Vector Machine (SVM).The main goal of this study is to extract the features of the different types of banknotes for efficient recognition.The color features were extracted using Histogram Equalization (HE) which are Red, Green and Blue (RGB) valuesin analysing the characteristics of five types of banknotes which are 100 Singapore Dollar, 100 Malaysia Ringgit, 100 Nepal Rupee, 100 Bangladesh Taka, and 100 China Yuan.The performance of the proposed study is evaluated to 150 testing images which produced97.33% of overall mean recognition accuracy.It is believed that the study outcome could assist the financial institution particularly to recognize various types of banknotes conveniently and efficiently.