Fake Money Detection using Machine Learning

Our nation's most valuable asset is bank currency, and in order to cause financial inconsistencies, counterfeit notes that seem like the real thing are introduced into the financial market. A significant amount of counterfeit money was observed to be floating on the market during the demonetization period. Because many characteristics of a forged note are identical to those of an original, it is generally exceedingly difficult for a human to distinguish a forged note from a genuine one using the numerous factors intended for identification. It is difficult to distinguish between authentic and counterfeit banknotes. Therefore, an automated system must be provided in ATMs or banks. An effective algorithm that can determine if a banknote is authentic or counterfeit must be created in order to create such an automated system, since counterfeit notes are quite precisely manufactured. In order to detect bank cash authenticity, we use the CNN method in this research on datasets that are accessible through the UCI machine learning repository. We have used machine learning methods to accomplish this, and their performance is evaluated using a variety of quantitative analysis parameters. Keywords: counterfeit currency, fake money detection, machine learning, image processing, Convolutional Neural Networks, Support Vector Machines, automated systems.