Designing a Model to Monitor the Commitment of People to Wear a Face Mask, Even if an Image of their Real Face is Printed on it, using Deep Learning

The novel coronavirus (COVID-19) has rapidly affected our daily lives, which led to stopping the wheel of life for millions of people around the world and disrupting trade and global movements. The most important measure to limit the spread of this virus is to wear a mask on the face, so this procedure has become a new normal. In the near future, many public service providers will require customers to wear masks properly to benefit from their services. Therefore, the discovery of the face mask has become a critical task to help the global and local community. In this paper, we propose a method using deep learning techniques and models that detects the faces of several people in a image or video stream in real-time and then identifies the faces of people who wear a mask and people who do not wear a mask, it also discovers people who put a mask in the wrong way, that is, the mask is under the chin or covers the mouth and chin only, and the nose appears, and here it means its absence, and also discovers people who wear a mask with the real features of their faces hidden behind the mask, that is, they put a mask but appear as if they are not wearing a mask.