Abstract
Technological advancement had replaced humans with machines in almost every field. Banking automation have reduced human workload by introducing machines. Tedious task like currency handling that require more care are simplified by banking automation. When machines are handling currency they should recognize it. In this paper a method for currency recognition using principal component analysis is implemented. Principal components of currency features are extracted and weight vector is computed for the same. The weight vector similarities are then computed using Mahalanobis distance measure. For prediction the image having least distance measure with a class is determined. We observed that both the central numeral feature and RBI seal could classify the unknown currency with 96% accuracy. Thus our proposed currency recognition system can be integrated with the currency sorter of ATM machines.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
