Counterfeit Currency Detection on Indonesian Banknotes Based on Latent Image Security Feature
Andrean Limanto · International Journal of Advanced Trends in Computer Science and Engineering · 2020
The advancement in printing technologies is in line to the advancement in money counterfeiting technologies, especially in Indonesian banknotes.The authorities are trying to protect the Indonesian banknotes from counterfeiting by embedding up to 12 security features on the banknotes.By its characteristic, the security features are divided into three, which are open, semi-open and closed.However, with so many security features, counterfeit currency on Indonesian banknotes still circulating because the counterfeiters can make a banknote that is very similiar to the original banknote.But, the counterfeiters pay less attention to the details of latent image security feature.Latent image is a semi-open security feature that must be viewed carefully to be able to recognize it.Although this feature can be printed and falsified, the quality of this feature on conterfeit banknotewas very bad.Based on this, original banknote and counterfeit banknote can be detected and classified according to the latent image security features.This research aims to develop a method for detecting the authenticity of banknotes using Deep Learning technology with the Convolutional Neural Network (CNN) architecture.Outcome of the research can be used to develop an expert system for counterfeit banknotesdetection on Indonesian currency.