A Real-Time Face Mask Detection Using SSD and MobileNetV2

Abstract

After a rapid spread of Coronavirus (COVID-19) in Wuhan-China in December 2019, the World Health Organization (WHO) confirmed that this was a dangerous virus that could spread from person to person through droplets and airborne contaminants. To prevent the spread of the Covid19, people should wear a mask during the epidemic. During this pandemic, it is becoming increasingly difficult to keep track of human beings the one who wears a mask as a usual practice or not. It will not solely depend on human efforts to keep track the whole world so there is a need to build software that automatically detects whether people in public places wearing a mask or not. Many new models are developed utilizing convolutional Neural Network to build a model as accurately as possible. The method proposed in this paper uses the ResNet model to obtain multiple faces with a single (SSD - Single Shot Multibox Detector) image using a network (model) and MobileNetV2 Architecture used as face mask detectors. This proposed model has 99% more accuracy than most other face recognition models. This mask detector model uses a dataset of hidden morphed masked images to obtain more accurate model. This system should be used in Real-time applications which require face mask discovery for protection purpose due to the sudden happening of Covid-19.