Classroom Automation System Using Human Detection Based on YOLO Algorithm and IoT Technology

Classroom is an important facility in the school or university to be utilized as a learning space and lessons happen. Students and teachers meet in the classroom on a daily, with different schedules. Hence, every class needs to be monitored and controlled for finding out the class conditions. However, the staffs have to monitor the classroom by checking the classroom one by one. Thus, it is not efficient and increase the workload if there are many classes to be checked. Therefore, automated technology can be implemented to reduces the workload and increases staff efficiency when controlling and monitoring the classroom, especially in the context of time. In this study, an IoT-based system was proposed as a solution to control the classroom condition. This system used class schedules and human presence as a parameter to control the classroom. A camera utilized to detect the object and performed YOLOv8 algorithm to analyze the condition. in this system, firebase was leveraged to store every data from detecting process. Then, the data was used for controlling the actuators, such as door locking, lamp, and fan, which were controlled by ESP32. The result showed that the proposed system could perform with accuracy of 57%, sensitivity of 67%, and positive predictive value of 79%. The proposed system can achieve a success rate of 90% with less than 20 second of time delay which makes this system time effective. But it is not yet known whether this system will work effectively in low light conditions.