IDENTIFYING COUNTERFEIT NAIRA NOTES USING RECURRENT NEURAL NETWORK

Abstract

The naira functions as Nigeria's official currency to facilitate trade and investments while serving daily transactions. 
The issue of unaccepted Naira notes creates genuine problems which bring national disgrace since it reduces 
economic speed and weakens trust in banking institutions. The cash system suffers integrity damage while public 
trust in the economy declines causing major financial losses to individuals and both emerging and established 
financial institutions. The following report explains how to construct and train a Recurrent Neural Network (RNN) 
model for counterfeit Naira notes detection. The implementation of Deep learning techniques and image detection 
methods generates exceptionally high detection accuracy in this work. The selection of Histogram of Oriented 
Gradients (HOG) as a feature extraction technique play a vital role in developing a system that properly identifies 
original notes from counterfeits. The accomplishment addresses Nigeria's economic challenge of counterfeit money 
by solving this significant problem.