Student Surveillance System using Face Recognition

This study aims to enhance campus security and real-time monitoring through the development of an intelligent surveillance system. Computer vision and face recognition technologies, specifically the Haar-cascade algorithm implemented using OpenCV are utilized. The system captures images from multiple cameras at various checkpoints and employs face detection and feature extraction algorithms to identify individuals. By comparing the detected faces with a database of authorized individuals, unauthorized individuals can be promptly identified. The objectives include achieving improved accuracy, real-time tracking, and proactive security measures. While the system demonstrated reliable performance under normal lighting conditions, challenges were encountered in low-light situations, leading to reduced face detection and recognition accuracy. These limitations were primarily attributed to the camera quality and the processing power of local machine. To overcome these limitations, future improvements will involve integrating higher-quality security cameras and more powerful hardware. These enhancements aim to optimize the system's performance and scalability for effective deployment in real-world scenarios. The integration of computer vision, machine learning, and IoT technologies in this study highlights their transformative potential in surveillance systems. In summary, the intelligent surveillance system successfully utilizes computer vision and face recognition techniques for campus security. Despite the limitations encountered, the system demonstrated promising results in real-time monitoring and identification with high accuracy.