Research On Paper Currency Recognition By Neural Networks

The technology of currency recognition is used to research the visible and hidden currency characters, identify features all-around and dispose of the process on time. The original information has a loss because paper currency will get to be worn and blurry, even damaged by human being in circulation. Furthermore the complexity of background designs of different kinds of RMB makes it difficult for pattern recognition to work well to the currency recognition. So it is an important guarantee for improving the recognition ratio to currency paper how to extract the characteristic information from currency image with noises and select a proper pattern recognition arithmetic.Now the main methods for paper currency recognition have geometrical dimension measure, feature recognition of image texture and NN etc. NN which is nonlinear math mode is used widely in the application field of currency recognition incorporating with some optimizing arithmetic such as genetic algorithm and simulated anneal in order to optimize NN structure and improve NN recognition precision. But how to extract characteristics information effectively and decrease the influence of noises to advance NN recognition ratio need more research. This article, aiming at the specialties of RMBcurrency image, puts forward a new method using linear transform of image gray to diminish the influence of the background image noises in order to give prominence to edge information of the image. Then the edge characteristic information image is obtained by edge detecting using simple statistics. By dividing the edge characteristic information image in the width direction into different areas, getting the number of the edge characteristic points of different areas as input vectors to random masks and optimized by GA. Finally the vectors optimized by GA are as the input vectors to NN, carrying out classifying recognition by three layer BP NN, paper currency is recognized. By experimental tests, recognition ratio to the new printing style of 100 RMB, the old printing style of 50 RMB, the new printing style of 50 RMB, 20 RMB, the new printing style of 10 RMB is 100%, 100%, 100%, 100% and 100% respectively. And the results are satisfying. This method has satisfying results by experiments and computer simulation. The proposed method of extracting input vectors has advantages of simplicity, high calculating speed, obvious characters of original image, good robuest to different kinds of RMB.