Precocious Fiat Currency Computing Robot

Counting and classifying currency notes with accuracy and high speed is a challenging problem for banks, stores, and even consumers. Before the arrival of currency counting machines, people used to count currency manually. This procedure was not only tedious and time-consuming but also prone to mistakes. Although relatively common in North American and European markets, counting machines have yet to be fully developed for the Indian market. This is partly because of the lack of distinctive features to differentiate between denominations and the variability in the features for a given denomination. This mini-project aims to see whether machine vision techniques can be used to recognize and count Indian currency notes accurately and at high speed. Although there are currency recognition algorithms documented in the literature, the applications work offline with static images of the notes. In this proposed mini-project, a color sensor-based system for the online classification of moving notes is evaluated.