Detection of Face Mask in Real-time using Convolutional Neural Networks and Open-CV

The world has witnessed a major uproar in the year 2020 with the widespread transmission of COVID-19. The propensity for wearing a face mask has become essential as a countermeasure against the transmission of the infection particularly in open settings where keeping up with social distancing is not functional more often than not. As a result, wearing a face mask is crucial to curb the spread of the pandemic. This paper is concerned with a simplified approach for the detection of face mask in real-time using convolutional neural networks (CNN) and Open-CV. The proposed CNN model trains on a dataset of 12000 images of faces with and without mask using two convolutional layers and predicts on real-time video streams using the Haar cascade classifier of Open-CV. The model reported an accuracy score of 98.8% on the training set and 99.37% on the validation set using the CNN architecture without undergoing any problems of overfitting.