Recognition of Indian Currency Denomination for Visually Impaired People

Currency denomination recognition is one of the basic tasks in our day to day transactions. However, it is difficult for blind people to identify the currency. Therefore automated recognition of currency denomination system would be of great significance to relieve their agony. In this paper a robust framework for Indian currency note recognition which would effectively recognize the various Indian currency denominations (10, 20, 50, 100, 500, 1000) thus benefiting the visually impaired and also counterfeit detection of currency. Here firstly preprocess the image of currency notes acquired by a digital camera and then extract and analyze three unique features of each denomination while considering complexities caused by degradation of color and dimension of notes due to wear and tear and handling. Based on these unique and discriminating features, classify and recognize of these currency notes can be done. The experimentation results showed good classification results and the success-rate of the counterfeit detection with properly captured image is 100%. An audio feedback is given for every test case scenarios.