BANK: A Banknote Acceptor with

Neural Kernel ?

A. Frosini??, M. Gori?, P. Priami??

? Dipartimento di Sistemi e Informatica, Universit?a di Firenze

Via di Santa Marta, 3 - 50139 Firenze - Italy

Tel. +39 (55) 479.6265 - Fax +39 (55) 479.6363

E-mail: [email protected]

WWW: http://www-dsi.ing.unifi.it/~marco
?? Consulting Engineer

Abstract

This paper gives a summary of the electronics and software modules of BANK, a banknote machine operating by a neural network-based recognition model. The machine perceives banknotes by means of low cost optoelectronic devices which produce signals associated with the reflected and refracted rays of two parallel strips in the banknote. The recognition model is based on multilayer networks acting for both the classification and verification steps.

1 Introduction

Automatic currency recognition is increasing its importance in the distribution of different products ranging from cigarettes to bus tickets and also in automatic currency change. Therefore, automatic machines capable of recognizing banknotes are massively used in automatic dispensers as well as in a number of automatic banking operations.

The wide range of applications of these machines places severe constraints on their cost, performances, and, consequently, architectural solutions. For instance, most machines operating in automatic dispensers are not very effective in terms of performance but, on the other hand, they are available at very low cost. Whereas machines used in banking operations usually have to pass very severe tests, but are available at higher cost.

Basically, humans recognize banknotes quite effectively, albeit they may be betrayed by good imitations, particularly when they do not look at the banknote carefully. The typographic techniques for imitation are becoming more affordable than in the past and, therefore, this is playing more serious constraints on the design of effective banknote acceptors. Humans recognize banknotes on the basis of a combined check: they inspect the picture as well as the paper consistency and, sometime, they also look for special detection signs

?BANK project has been managing by DF Elettronica and DF Sistemi, Calenzano (Italy).