Facemask Wearing Detection Based on Deep CNN to Control COVID-19 Transmission

Abstract

Since the expansion of the COVID-19, almost all countries have advocated their residents to put on facemasks and adopt social distance and hand cleanliness. Due to the complicated attitudes in the settings of real life, besides several socio-behavioral and cultural factors, it is not easy to give a convincing situation for the general public that wearing facemasks is useful and effective. Therefore, facemasks wearing has not been widely embraced by many residents. However, the usage of facemasks has offered the considerable potential to filter or block the transmission of respiratory viruses including COVID-19. In this paper, a model of deep convolutional neural network (CNN) for facemask wearing detection is proposed to control covid-19 transmission. This proposed deep learning model includes two main processes; feature extraction and classification. The CNN classifier provides 99.57% of accuracy for the utilized Real-World Masked Face Dataset (RMFD).