Real-Time COVID-19 Infection Risk Assessment And Alert System Using Mask Detection

The onset of COVID-19 pandemic has led to many life-style changes, among them is the usage of Face masks to protect themselves from the deadly virus. It is recommended to use face masks, but it is a common practice for people to forget wearing it. Sometimes both the nose and mouth are not covered leaving them at a risk of infection. The purpose of this paper is to alert shopkeepers or property owners about the risk of spread of Covid virus due to the non-usage of face masks. By monitoring a place for number of people wearing and not wearing face masks, one is able to be aware of potential risks people carry. The owner is alerted about the count of non-masked people when it reaches a certain threshold through a siren and also through an alert message sent on a popular messaging platform. This way, one can persuade people to wear masks or to leave the premises, thereby protecting himself and others. This paper uses VGG Face model, OpenCV and TensorFlow and a custom trained mask detection model to achieve the same.