Face Mask Detector Using Machine Learning Applications

Abstract

Wearing face masks is required for everyone during this pandemic because the COVID-19 virus may be transmitted via the mouth, nose, or eyes, which can occur when a person comes into direct or close contact with someone who has the virus [1]. However, despite the stringent implementation, some individuals neglect the correct use of face masks, oblivious to the hazards of probable virus transmission. This paper will show how a Convolution Neural Network (CNN) can detect if a person is wearing a face mask or not, as well as an additional parameter to support detecting if the face mask is properly worn by a person by considering facial landmarks via face recognition using a linear SVM machine learning algorithm and a Histogram of Oriented Gradients (HOG) feature descriptor [2]. The detection of correct face mask usage involves two (2) stages. Face Mask Detection must pass in order to progress to the next procedure, Face Detection, where the result of checking must return false to validate the appropriate wearing of a person's face mask [3].