Real-Time Student Activity Detection and Incident Monitoring Using Artificial Intelligence

The shift towards offline education has made it challenging to monitor and supervise student behavior effectively, it is more difficult to keep an eye on their behavior, which leads to disruptions and diversions during classes. Although several algorithms recognize these behaviors, their efficiency and accuracy are constrained and mostly target individual items. In response, we propose an innovative student tracking system driven by artificial intelligence (AI), which will transform behavior analysis, attendance control, and incident detection in educational institutions with real-time monitoring. To record real-time insights regarding student behaviors like napping in class, using mobile phones, and engaging in irregular activities, the system uses AI techniques like Convolutional Neural Networks (CNNs), OpenCV, Face Recognition Module, and YOLOv8. These techniques are also integrated with real-time alerts using the Twilio module. Using carefully placed cameras, the main process is image recognition; the AI model is instructed to identify and classify various activities quickly. The YOLOv8 architecture, which can record real-time video at a higher Frames per Second (FPS), is used to train the model. The system evolves through continuous development, ensuring that it remains successful in addressing changing educational settings and dynamic student behaviors