Currency Detection for Blind Person Using CNN
Anurag Kumar, Aman Kushwaha, Yuvraj Singh · International Journal of Research Publication and Reviews · 2024
In today's world every day thousands of people get scammed because of fake notes.For a normal person, it's not that much of a problem to detect the difference between a fake and a real note, but in the case of visually impaired persons, it's not the same as they can't see the minor differences.In this project we are using HOG (Histogram of Oriented Gradients) and Region-based Convolutional Neural Networks (R-CNN) to create an AI/ML model.Even though special symbols are stamped on various denominations in India, the task is still tedious for the blind.The lack of identification devices motivated the need for a hand-held device to segregate the different denominations.In this project, the features of the images are compared with all the reference images of the currency, if the difference is less than a threshold value, the numerical part of the currency is extracted and compared, if it matches, the corresponding denomination of the currency is recognized..We are using HOG because it has a few key ad vantages over other descriptors.Because it operates on local cells, it is invariant to geometric and photometric transformations, except for object orientation..The key reason behind using the R-CNN series is region proposals.