Paper Currency Recognition System Using Characteristics Extraction and Negatively Correlated NN Ensemble

An efficient currency recognition system is vital for the automation in many sectors such as vending machine, rail way ticket counter, banking system, shopping mall, currency exchange service etc. The paper currency recognition is significant for a number of reasons. a) They become old early than coins; b) The possibility of joining broken currency is greater than that of coin currency; c) Coin currency is restricted to smaller range. This paper discusses a technique for paper currency recognition. Three characteristics of paper currencies are considered here including size, color and texture. By using image histogram, plenitude of different colors in a paper currency is calculated and compared with the one in the reference paper currency. The Markov chain concept has been considered to model texture of the paper currencies as a random process. The method discussed in this paper can be used for recognizing paper currencies from different countries. This paper also represents a currency recognition system using ensemble neural network (ENN). The individual neural networks in an ENN are skilled via negative correlation learning. The purpose of using negative correlation learning is to skill the individuals in an ensemble on different parts or portion of input patterns. The obtainable currencies in the market consist of new, old and noisy ones. It is sometime difficult for a system to identify these currencies; therefore a system that uses ENN to identify them is discussed. Ensemble network is much helpful for the categorization of different types of currency. It minimizes the chances of misclassification than a single network and ensemble network with independent training.