Abstract
Paper currency recognition (PCR) is one sort of insightful framework which is a significant need of the present computerization frameworks in the cutting edge universe of today. It has different potential applications including electronic banking, cash checking frameworks, cash trade machines, and so on. This paper proposes a programmed paper cash acknowledgment framework for paper currency. It utilizes InceptionV3 for extraction and neural network for classification and utilizes the instance of Indian paper currency as a model. The strategy is very sensible regarding exactness. The framework manages 3100 pictures. The pictures are dispersed in six classifications—10, 20, 50, 100, 200, and 500, and they are being utilized for examination and characterization. The proposed calculation is completely programmed and requires no human intercession. To approve the adequacy of system and appropriateness of neural network for cash picture arrangement, examinations have been finished with different classifiers like K-nearest neighbors (KNN) and support vector machine (SVM). The proposed procedure delivers very palatable outcomes as far as acknowledgment and effectiveness as it has accomplished an average accuracy of 97.2% over six categories. KeywordsPaper currencyImage processingNeural networkInceptionV3Content-based classification
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have