FAKE CURRENCY DETECTION USING MACHINE LEARNING
Mrs.J.Maheswari Assistant Professor, Computer Science and Engineering , Dhirajlal Gandhi College of Technology
Mr.R.Vijay Student, Computer Science and Engineering , Dhirajlal Gandhi College of Technology
Mr.K.Yogesh Kumar Student, Computer Science and Engineering , Dhirajlal Gandhi College of Technology
Mr.P.Sudhagar Student, Computer Science and Engineering , Dhirajlal Gandhi College of Technology
Mr.K.Vishnu Student, Computer Science and Engineering, Dhirajlal Gandhi College of Technology
---------------------------------------------------------------------***---------------------------------------------------------------------
I.ABSTRACT
As technological advancements and scientific research have improved our daily lives, human-computer contact has grown to become a need. The use of these technologies will enable people who are blind to participate in some social activities. Therefore, this initiative has been started as a good start for the blind people in order to help them blend with their surroundings and society and also to be independent in conducting their daily routine chores. In order to enable blind persons to easily browse or use the gadget's functionalities to interact with others in society, there should be an assistive device for the visually impaired. For visually impaired people, cash recognition and fraudulent note detection is an effort to improve living conditions for blind people. Deep learning has overtaken all other study areas in recent years in popularity. The dataset is primarily trained using neural networks. This research endeavor can make use of a wide variety of models. Correctness of currency recognizing can be increased using these models. Such study techniques are, of course, consistent with what we would expect. In this study, we primarily use the Single Shot Multibox Detector (SSD) model, which is based on deep learning, as the framework. We then use the Convolutional Neural Network (CNN) model to extract the characteristics of paper money, allowing us to more precisely identify the denomination of the cash on both the front and back.
Key Words:
Convolution neural network, Currency detection, Deep learning, Feature extraction, Image processing.
