Summary

Smart Info-Media Systems in Asia

2019

Session Number:SS2

Session:

Number:SS2-3

Alphabet and Number Recognition of Banknote by Using the Deep Learning

Daichi Kayanuma,  Akira Tehara,  Hiroyuki Tsuji,  Tomoaki Kimura,  

pp.34-37

Publication Date:2019/9/4

Online ISSN:2188-5079

DOI:10.34385/proc.57.SS2-3

PDF download (1.3MB)

Summary:
The forgery of banknotes has been around for a long time. It is effective to recognize the banknote number as a way to confirm the forgery of the banknote. However, the conventional methods such as OCR are not suitable for recognition of identification numbers. Therefore, an approach using deep learning is tried. In this paper, we verify the recognition accuracy of the unverified alphabet and number, and also change the way to give the training data and confirm whether the recognition accuracy changes. As a result, a recognition rate of 96% or more was confirmed.