Bank Locker Authentication System using Facial Recognition

Abstract

Security and Authentication of individuals is necessary for our daily lives especially in Bank lockers. A smart digital door lock system for bank automation is equipment that uses digital information such as a user’s data and face recognition as the method for authentication in the system. In this system the bank will collect the face imprints of each person for accessing the lockers. Only authenticated people can recover the money, documents from the lockers as faces are stored for the individual identity of a person. For facial recognition, this project uses the CNN algorithm. In this project, only authenticated user can access the lockers as faces are stored for the individual identity of a person. Facial recognition alone cannot determine whether the person is real or not. Therefore, liveness detection is implemented. In liveness detection, the system detects if it interacts with a real person or a spoof artefact used by other person such as a face photo. To detect whether the person is live or not the project uses eye blink detection. In this system the owner of a particular locker needs to authenticate his/her identity by verifying their face following with an One Time Password authentication feature. Here the person will receive an OTP received on their registered mobile number, proceeding with it; the valid one will allow the owner to access the particular locker whilst incorrect will not.