ATM System using Biometrics

Abstract

Today, money transactions are done using ATM cards which play a key role in the nature of trade. The problem with the design of the currently used ATM Systems which includes password & PIN is that they do not have an authentication system to find an authorized user & they also do not provide any security mechanism in case of card misuse. We gained initial ideas by researching current ATM Systems to come up with this design. We then took what we found and made the most user-friendly product possible. For this project we created an ATM System using Biometrics. In this system the user will first have to authenticate himself using various Biometric techniques such as Face Recognition, Fingerprint Recognition, OTP generation using Voice Commands & finally by entering the PIN. We’ve also added a ML component to give it an upper hand. The system will undergo learning about the number of transactions, transaction amount & transaction location & will accordingly enable the security mechanism in case the usual limit exceeds, such as freezing the account for a particular time-period. For communication purpose we’ve used Kerberos which makes use of the third-party utility to authenticate the client to file server. In case of an exception, if an authorized user is unable to access his/her account then in that worst-case scenario, the user will use voice automated code word to login to his account.