Assessment of Smart Campus Surveillance and Guidance Approaches

Abstract

The increasing size and complexity of modern campuses has created challenges in ensuring safety, security, and seamless navigation for students and visitors. Traditional manual surveillance methods and static campus maps are inadequate in providing real-time monitoring and dynamic guidance. Human-dependent monitoring is error-prone and inefficient, while students and visitors often face difficulties in locating classrooms, administrative offices, or other facilities. Moreover, issues such as inconsistent attendance tracking, class bunking, and unidentified anomalies further highlight the limitations of current systems.

To address these challenges, this project proposes a software-only, machine learning–based Smart Campus Surveillance and Guidance System that leverages existing CCTV infrastructure. The system integrates face recognition for automated attendance tracking, anomaly detection for suspicious activities, and an indoor navigation module for real-time campus guidance. The solution operates without the need for additional IoT or blockchain devices, making it lightweight, cost-efficient, and scalable.

The architecture comprises multiple software modules including preprocessing, face recognition, activity detection, and navigation built on graph-based pathfinding algorithms. Backend APIs and a database store all student, attendance, and anomaly data, while a user-friendly frontend provides real-time alerts and navigation. By combining deep learning models such as YOLO, FaceNet, and LSTMs with pathfinding algorithms like Dijkstra and A*, the system ensures intelligent monitoring, early anomaly detection, and smooth user navigation.

This system has the potential to revolutionize campus security and student management by automating critical tasks, reducing manual errors, and improving campus safety standards. Beyond surveillance, it also enhances user experience by providing personalized route guidance and timely notifications. With privacy safeguards, scalability features, and modular deployment, this project lays a strong foundation for the adoption of AI-driven smart campuses in the future.