Serial Number Extracting and Recognizing Applied in Paper Currency Sorting System Based on RBF Network

In view of the increasing demand of recognition system for paper currency number, to develop a type of number recognition system based on CIS and DSP. The hardware is composed of CIS and DSP which control image acquisition and process. The software is composed of image acquisition, character correction and recognition. To recognize character with noise pollution rapidly and accurately, a novel approach for character feature extraction based on statistics and fuzzy membership is proposed, and applied in RBF neural network. The experiment shows that the recognition system runs stably and accurately, the collected image shows clearly. The system runs with highly exact rate on number recognition.