Fake Currency Note Recognition using Extreme Learning Machine

Abstract

Color printing technology has made it simpler for forgers to make and copy large numbers of fake currency. The production of legible banknotes was once the sole purview of print shops, but today even a novice printer with a simple laser printer can do the job. Because of this, the prevalence of counterfeit money has increased dramatically. A number of problems, including corruption and the existence of “black money,” have hounded India for a long time. There is also a serious problem with the circulation of counterfeit currency. The outcome is a system that can detect fake currency with greater speed and precision. It is a serious crime to create counterfeit currency, which is money that has not been authorized by the government. The vast majority of them do this professionally. The circulation of counterfeit currency retards economic development and devalues legitimate currency. In this study, it uses image processing to the problem of identifying fake Indian notes. This technique finishes the initial acquisition before the image is processed further. Pre-processing includes actions such as cropping, smoothing, and modifying the image; converting the image to greyscale; applying image segmentation; extracting features; and resizing the image. The model is then trained using an ELM. Proposed model exceeds both the CNN and SVM models in terms of accuracy, at around 97.8 percent.