Face Mask Detection and Alert System Using Convolution Neural Networks

ABSTRACT

The present situation of the COVID-19 pandemic called for very strict safety norms to contain the virus. One of these mainly included wearing a mask at all places outside one's home. However, many do not take that seriously. This paper deals with an approach to detect a face mask on an individual's face, which if violated can be used in an alert system, and the like. The algorithm used to train the machine is convolution neural networks (CNN) for face(image) recognition in real-time. The results and inferences obtained are with respect to the model structure, algorithm, and variation in the dataset. The final model had an accuracy of 96.18% and was stable in real-time with no false-negative results. The final application alerts the concerned if the mask is not worn properly and captures an image to be sent to the concerned. © 2021 IEEE.