Smart Library Occupancy Management System using Deep Learning

Abstract

Libraries play a crucial role in providing a place for students to gain knowledge. However, in today’s world, library spaces are often crowded, and managing real-time occupancy is challenging. Smart Library Occupancy Management System is a web-based system which is mainly designed for Perpustakaan Tunku Tun Aminah (PTTA) to manage the library's occupancy. This system will detect the person entering and exiting and provide real-time occupancy data. This system uses deep learning algorithms, specifically YOLO (You Only Look Once) for person detection, YuNet for face detection, and MobileFaceNet for face recognition, to monitor real-time occupancy in the library. YOLO is a fast and accurate object detection model and YuNet is a model for face detection and capable of identifying facial landmarks with high precision. MobileFaceNet is a lightweight convolutional neural network optimized for face recognition tasks, offering a good balance between accuracy and computational efficiency, making it suitable for real-time applications. Additionally, it includes a web-based room reservation system, allowing students to check room availability and make reservations online. After testing, the system able to manage library occupancy using deep learning algorithms and future work could be focused on increasing the facial recognition accuracy under different circumstances.