A Check on WHO Protocol Implementation for COVID-19 Using IoT

The lockdowns imposed by the governments worldwide is proving to be one of the most taxing repercussion of COVID 19 pandemic. One of the efficacious precautionary methods recommended by the World Health Organization (WHO) is to wear a face mask when in close contact with other people. In this chapter, Raspberry Pi 3 has been used to implement a TensorFlow-based Deep Learning Neural Network model that tries to recognize if a person is wearing a face mask or not using Raspberry Pi camera module. It also notes his/her body temperature using a non-contact infrared temperature sensor. The proposed solution is mitigating the problem where people casually avoid wearing a mask on their face by detecting face mask and measuring body temperature which should fall in the normal human body temperature range simultaneously. Various examples have been used to train the model to make it robust. The dataset consisted of 1509 images in which 755 images were with mask and 754 images were without mask. The proposed deep learning model has an accuracy of 97.8% for face mask detection after training on TensorFlow GPU 2.0.0 which is best achieved till date as per authors’ knowledge.