Enhancing Campus Security: Face Detection and Automatic Recognition System for Guelma University

Abstract

In this thesis we present the different approaches to face detection and feature extraction, as well as approaches to facial recognition. The study introduces a real-time facial recognition system designed for enhancing security in institutional access points. The system can recognize faces in real time even when there are many faces at once. This is achieved through the use of advanced computer vision and deep learning algorithms that detect people as they walk into an institution. By comparing the identified faces with a database already established by security personnel, the system can immediately alert them about any unauthorized or unknown person trying to enter. The system also records the entry and exit times of recognized individuals. To achieve better face detection, our study focuses on MTCNN based face detection technique. First, the output of face detection will then be subjected to feature extraction using Inception-ResNet-v1 for identification. the system demonstrated good performance in high accuracy real-time monitoring and identification