A System Design With Deep Learning and IoT to Ensure Education Continuity for Post-COVID

Abstract

In this paper, an IoT and deep learning-based comprehensive study to reduce the effects of COVID-19 on the education system is presented. The proposed system consists of an edge device, IoT nodes, and a neural network that runs on a server. The purpose of the proposed system is to protect students and staff against infectious diseases and increase the students performance during classes by monitoring the environmental conditions via an IoT-based sensor network, during the current pandemic to ensure the use of masks in closed areas by training a customized deep learning model, and to monitor the student attendance data by deep learning and IoT-based solution. Furthermore, effective heating and cooling can be done to save energy by transmitting the environmental conditions of the indoor environment to the relevant destinations. The experiment is conducted with five different networks to classify the faces in the images as masked or unmasked, and their performances were examined. The networks were trained on the Face Mask Detection Dataset which contains a total of 7553 masked and unmasked images. The best results were obtained as 99.5% for the F1 Score and 99% for MCC by the model trained on the InceptionV3 network.