Identification of currency with artificial intelligence techniques

Counterfeiting of paper currency is a major issue around the world. Almost every country has been severely impacted by this, which has escalated into a major issue. The major goal of this research is to identify Indian paper currency. We acquired a dataset of the currency notes for both original and fake notes of different denominations from internet.

By using feature extraction method of front side of the currency to classify whether the currency is original or fake. In this paper we used the Support Vector Machine (SVM) algorithm for the classification. Perceive whether something is Original or fake. We have used the MATLAB image processing toolbox.

Image processing is a method of enhancing an image's visual information for machine or hardware perception. And to classify the denomination of the note we used a CNN network and the average accuracy of the trained network is around 90% with minimal error.

Keywords: - Image Processing, Monetary Identification, Denomination, Currency Identification, Machine learning, SVM, Brisk Features, Convolutional Neural Network.