Real-Time Facemask and Body Temperature Detection for Covid-19 Pandemic

Abstract

Population across the globe is striving against the COVID-19 pandemic so it is essential to prevent the spread of disease. In this pandemic situation, there is a fair chance of transmission of the virus by sputtering, therefore a face mask is necessary for the people to protect them from disease. Wearing face masks is the safety protocol that needs to be followed in public places to prevent the spread of the virus. Moreover, COVID-19 is also identified by fever. To differ between a healthy and unhealthy person we need to detect temperature in the real-time scenario. We are using deep learning libraries to detect face mask. Open CV is used for all sorts of image and video analysis, like facial recognition and detection. Open CV can process images and videos to identify objects and faces. Keras is a high-level, deep learning API developed by Google for implementing neural networks. It is written in Python and is used to make the implementation of neural networks easy. The frameworks supported by Keras is TensorFlow. Raspberry pi will help us to build such a system that will help us to detect temperature contactless. Convolutional Neural Network (CNN) used in this project help us to obtain accuracy of 97% approximately, the data is trained on real time dataset having a data of approximately 4000 images.