Enhancing School Safety and Management: A Face Recognition-IoT Based System for Attendance and Behavior Detection

This Ensuring high safety and management in schools is a critical concern for educators and administrators. In this paper, we propose a system that address this issue. Specifically, our system offers two solutions: an attendance system that uses face recognition techniques to automatically record attendance and a human recognition system that can detect abnormal or dangerous behavior in school environments, with a focus on fighting action. The attendance system uses face recognition technology to identify students and staff as they enter the school premises. The system records the attendance data and sends it to a central database using Message Queuing Telemetry Transport (MQTT). This approach reduces the time and effort required for manual attendance taking, thus enhancing efficiency and accuracy. The human recognition system analyzes video frames from closed-circuit television (CCTV) cameras to detect any abnormal or dangerous behavior, such as fighting action, in real-time. We utilize Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) video classification techniques to achieve this task. The CNN is used to extract features from video frames, while the RNN is used to model the temporal aspect of the video sequence. The proposed system has the potential to enhance school safety and management. However, privacy and surveillance concerns must be carefully considered when deploying face recognition technology in school settings. By addressing these concerns and implementing appropriate safeguards, our system can provide a valuable tool for educators and administrators to ensure a safer and more secure school environment.