A deep learning and internet of things based system proposal to reduce the effects of infectious diseases on education

Since early 2020, the COVID-19 pandemic as been negatively affecting people's lives in many areas worldwide. The field of education has also been one of the areas most affected by this situation, with the restrictions on face-to-face education. In this research, a system is presented to reduce the effects of both COVID-19 and other different infectious disease outbreaks that may occur in the future on education. This system checks whether the students entering the classroom are wearing masks with deep learning methods. Five different convolutional neural network models were trained on the same dataset and compared to classify masked and unmasked face images. The system also performs facial recognition with computer vision techniques and takes attendance to prevent students who need to be in quarantine from entering the classrooms. Three different face detection methods were compared to make face recognition faster. Finally, with the help of the Internet of Things, the temperature, relative humidity, and air quality of the classroom are continuously measured. The sensor data transmitted to the main server with the help of the MQTT server is analyzed. A warning is given to end the lesson before the environmental conditions suitable for spreading the virus occur. With the system created, it can efficiently control whether restrictions are complied with during epidemic periods. In this way, it is aimed to reduce the spread of the disease in crowded environments such as schools without disrupting face-to-face education.