Practical Design Methodology for Commercial Automatic Coin Recognitzers based on Neural Decision Engines

摘要

This paper will address the applicability of artificial neural network models to real tasks in industrial environments. For this purpose, we shall present a general methodology which will be outlined by means of a case study which consists in the implmentation of a the classification/dcision engine to the included in an automatic coin classifier. This coin classifier is part of currently available commercial vending machines. The methodology presented can be considered as divided in three main tasks: database compliation, selection of the proper neural model and implementation. The experimental results provided demonstrate that the use of artificial neural network models overcomes some of the limitations inherent to the traditional techniques considered when solving this task.