| Hi-index | 0.00 |
In this paper a new Persian banknote recognition system using wavelet transform and neural network has been proposed. The required images for the selected banknotes are obtained using a scanner. The color images are first converted to gray scale images, and then the discrete wavelet transform (DWT) is applied on the selected images and features are extracted. Finally, a multi layered Perceptron (MLP) Neural Network (NN) is presented to classify eight classes of interest, which are 50, 100, 200, 500, 1000, 2000, 5000 and 10000 to man notes. The system was implemented and tested using a data set of 320 samples of Persian banknotes, 40 images for each sign (from both sides). The experiments showed excellent classification results. The system was able to recognize more than 99% of all data, correctly.