フーリエ変換を前処理とするニューラルネットワークによる紙幣識別

抄録

In conventional bill money recognition machine, we develop the recognition algorithm according to the transaction speed and difference of various specifications. However, development of the algorithm for the recognition has been based on the trial and error method. Many researchers have reported that neural networks are suitable for pattern recognition because of the ability of selforganization, parallel processing, and generalization. In this paper, we present, a new bill money recognition method with neural network and show the effectiveness of the present algorithm compared with the pattern matching. Furthemore, we transform bill money data by FFT into frequency domain. Data representation by FFT is more preferable to bill money recognition with neural network and we show that the recognition ability can be improved still more by introducing a new measure of reliability.