Fake Currency Detection using Clustering and SVM Classification

Abstract

Coins and note currency are widely used in our daily life such as vending machines, parking meters, telephone booths and so on. In addition to being used as currency, people enjoy collecting coins and notes as they usually have artistic value and can give a vivid insight to the social life in history. However, in recent years, a lot of illegal counterfeiting rings manufacture and sell fake coins and at the same time fake note currency is printed as well, which have caused great loss and damage to the society. Thus it is imperative to be able to detect fake currency. We propose a new approach to detect fake Indian notes using their images. A currency image is represented in the dissimilarity space, which is a vector space constructed by comparing the image with a set of prototypes. Each dimension measures the dissimilarity between the image under consideration and a prototype. In order to obtain the dissimilarity between two coin images, the local key points on each image are detected and described. Based on the characteristics of the coin, the matched key points between the two images can be identified in an efficient manner. A post processing procedure is further proposed to remove mismatched key points. Due to the limited number of fake currency in real life, one-class learning is conducted for fake currency detection, so only genuine currency are needed to train the classifier.