A Real Time Face Mask Detection and Health Status Monitoring using Deep Learning after Pandemic

Over the past two years, COVID-19 has spread over 200 countries, leading to pandemic proportions. Still, the virus is rapidly changing, the variants spreading through direct and indirect contact in places irrespective of the vaccine distribution. Many infectious diseases spread through droplets, and micro-droplets require face masks and other social distancing measures. This article proposed for pandemic circumstances called real-time face mask with health screening (RTFMHS) with two screening methods: (i) Internet of Things (IoT) node monitors a real-time intelligent face mask detection using multi-level high-speed augmented convolution neural network (MLHS-CNN), (ii) Individuals health indicators such as non-contact body temper-ature sensor and blood oxygen saturation are sending to fog cloud to manipulate and display the user health status. A face mask detection technique based on MLHS-CNN is proposed to determine whether a person is wearing a mask properly. User health is computed in real-time by an RNN running on the fog server to estimate the risk of infection spreading. The proposed method uses a lightweight IoT node and fog-based Deep Learning (DL) tools for data analysis and diagnosis. The experimental results outperformed previous research in terms of detection accuracy, precision, recall, and time complexity.