Identifying Facemask-wearing Condition Using Image Super-Resolution with Classification Network to Prevent COVID-19

Abstract

Abstract Rapid worldwide spread of Coronavirus Disease 2019 (COVID 19) has resulted in a global pandemic. Correct facemask wearing is valuable in infectious disease control, but the effectiveness of facemasks has been diminished mostly due to improper wearing. However, there have not been any published reports on the automatic identification of facemask wearing conditions. In this study, we developed a new facemask wearing condition identification method in combination with image super resolution with classification network (SRCNet) SRCNet), which quantified a three categories classification problem based on unconstrained 2D facial image images. The proposed algorithm contained four main steps: image pre processing, face detection and crop, image super resolution, and face mask wearing conditions identification. Our method was trained and evaluated on public dataset Medical Masks Dataset containing 3835 images with 671 images of no facemask wearing, 134 images of incorrect facemask wearing, and 3030 images of correct facemask wearing. Finally, the proposed SRCNet achieved 98.70% accuracy and outperformed traditional end to end image classification methods using deep learning without image super resolution by over 1.5 in kappa. Our findings indicate that the proposed SRCNet could achieve high accuracy identification in facemask wearing conditions , which have potential application in epidemic prevention involving COVID 19.