ARTIFICIAL NEURAL NETWORK TO RECOGNIZE AN INDIAN CURRENCY NOTE USING UNIQUE IDENTIFICATION MARK

Abstract

Artificial neural network has a vast application and has been successfully applied to a broad spectrum of data intensive application such as financial, data mining, medical, operational analysis, industrial, science etc. This increase in application is due to its ability to solve problem where the relationship are quite dynamic or non linear. Therefore in this paper we have used ANN to recognize Indian currency note using one special feature of Indian currency note known as Identification Mark (I.D mark). The I.D mark is of different shape for different currency note except for Rs 10 where there is no identification mark present. In the proposed method first we define a window size based on the common region where there is I.D mark. Then based on the window size we have segmented the I.D mark from the currency note. After this Fourier Descriptor is used to extract the feature from the segmented portion. Then the identification of this extracted feature is done by using ANN.