Smart Transaction through an ATM Machine using Face Recognition

Abstract

Automated Teller Machines (ATMs) are a convenient way for people to access their cash and banking services. However, traditional ATMs are susceptible to fraud, such as card skimming and PIN theft. Face recognition technology has the potential to improve ATM security by providing a more secure and convenient way for users to authenticate themselves. This research paper proposes a system for smart transaction through ATM machines using face recognition. To proceed transactions user, need to enter their registered mobile number on which an OTP will be generated subsequently the camera mounted on top of the ATM will capture user’s face which will then be compared with the picture associated by the contact number using custom trained YOLO algorithm. Upon successful matching user need to verify OTP received and if verified they can access the ATM functionalities. The dataset used for this model i.e., Contact number and customer photograph will be collected by the bank at the time of customer account opening. The system was evaluated using a dataset of over 10,000 random face images. The results showed that the system was able to achieve an accuracy of over 99% in authenticating users. The system was also able to detect and prevent unauthorized access attempts. The proposed system has the potential to revolutionize the way that people use ATMs. It can make ATMs more secure, convenient, and fraud-resistant.