Identification of Unmasked Faces: Protection in Public Places from COVID-19 and Other Infections

Abstract

The $3^{\mathrm{r}\mathrm{d}}$ wave of Coronavirus is hitting the world. The doctors and scientists are also in their last phase of developing a vaccine. But it is challenging to get the vaccine for everyone due to several factors such as production, supply chain, cold chain and price. Till then, the only way to prevent infection is to wear a mark and follow social distancing. This paper describes how to identify mask and no-mask faces in public areas using image processing algorithms. One of the significant steps taken by people around the world to stay safe is wearing masks. This research-based project gets a camera-based solution to detect faces and recognize them if they exist in the database. If the live stream has unmasked faces, which aren’t present in the existing database, give out an alarm to alert the administrator or user. It is a significant project for times like these because now, with the lockdown ending in the world because of financial meltdowns, most public places like malls, temples, shops, grocery stores, and medical stores are opening. A concern for the safety and well-being of everyone around us because the virus can spread at a logarithmic rate and cause mass damage in the form of large bills and some cases, death. To avoid this, we always have to wear masks when in public places, and this is what the project does. The crucial technologies used in our manuscript are Face Detection, Face Recognition & Identification, and Feature Detection in the face.