Application of Computer Vision for Multi-Layered Security to ATM Machine using Deep Learning Concept

In today's world, the area of computer vision is advancing at a breakneck pace. Specifically, the goal of this study is to give a comprehensive overview of current breakthroughs in the use of computer vision methods to solve the security risk associated with accessing ATM machines. It is proposed in this paper that a complete solution for ATM security be developed, which would help in the enhancement of privacy and security via the use of emerging facial authentication technologies. During transaction initiation, this approach leverages facial recognition by obtaining a real-time picture of the client and then assessing the true perspective, where the recognized face is checked against ATM card owners' data stored in the database. Face recognition technology allows the system to recognize and identify each user separately, thereby turning the face into a password. This eliminates the possibility of fraud resulting from ATM card theft and copying. A detailed review of the use of machine learning in computer vision and its implementation on embedded hardware, as well as distinct security issues for ATM machines, is provided in this study.