Classification of the new and used bills using the spectral patterns of raw
time-series acoustic data (observation data) poses some difficulty. This i
s the fact that the observation data include not only a bill sound, but als
o some motor sound and noise by a transaction machine. We have already repo
rted the method using adaptive digital filters (ADFs) to eliminate the moto
r sound and noise. In this paper, we propose an advanced technique to elimi
nate it by the neural networks (NNs). Only a bill sound is extracted from o
bservation data using prediction ability of the NNs. Classification process
ing of the new and used bills is performed by using the spectral data obtai
ned from the result of the ADFs and the NNs. Effectiveness of the proposed
method using the NNs is illustrated in comparison with former results using
ADFs.