A Neural Network-based Model for Paper Currency Recognition and Veriication

This paper describes the neural-based recognition and veriication techniques used in a banknote machine, recently implemented for accepting paper currency of diierent countries. The perception mechanism is based on low cost optoelectronic devices which produce a signal associated with the light refracted by the banknotes. The classiication and veriication steps are carried out by a society of multi-layer perceptrons whose operation is properly scheduled by an external controlling algorithm, which guarantees real-time implementationon a standard microcontroller-based platform. The veriication relies mainly on the property of autoassociators to generate closed separation surfaces in the pattern space. The experimental results are very interesting, particularly when considering that the recognition and veriication steps are based on low cost sensors.