Title: "Bank Locker Security System Using Face Recognition "
         

Page(s): 77-81
Authors: Roshiny Thomas, Sanjana Mathews, Sona Ojus, Sona Roselin Joseph, Dhiraj K. Thote

Abstract

Abstract: This paper presents an advanced Bank locker security system based on Face Recognition. The locker holders will be able to access their locker after the system recognizes their face and exercises locker door control to allow entry. Detection is performed on the basis of skin color model. This will help reduce computational complexity. Viola Jones Algorithm is used for the same. An additional password protection is provided to enhance locker security. In case of a wrong password entry, the true locker holder will be notified with an alert call from the bank to the saved contact number in the system. The system is designed to capture a person’s real-time image, compare it with the stored images in the database and on detecting a match, will grant them access after password entry has been verified. Face recognition is performed using the Principal Component Analysis and a specific face can be recognized by comparing the principal components of the current face to those of the known individuals in a facial database built in advance. Illuminance variations are manipulated using the Discrete Cosine Transform Algorithm (DCT).

Index Terms— Face detection, Skin color, Face Recognition, Principal Component Analysis (PCA), DCT.