Design and Implementation of Multilayer Security for ATM Machines

Abstract

Automated teller machine (ATM) nowadays are a favourite spot for attackers as they are available everywhere and are much easier to rob. Generally, ATM attacks can be either physical ATM attacks or ATM-related fraud attacks. In this paper the idea of an ATM system with multilayer security is proposed with the help of internet of things (IoT), fingerprint identification and face recognition to increase the security of ATM. The physical ATM counter attacks can be identified by using specific sensors to detect changes in vibration and temperature in the ATM counter. To prevent ATM related fraud attacks the proposed system has additional security features like fingerprint identification and face recognition along with ATM number verification. The convolutional neural network (CNN) and machine learning based face recognition is used in this work which is quite reliable. Failures in any of the above steps cancel the transactions and so the proposed system provides multi layer security which makes it impossible for the attackers to break the ATM security. The proposed system will help to increase the security of the ATM and provide safe and secure ATM transactions.