Face Mask Detection and Face Recognition of Unmasked People in Organizations

With the onset of the global pandemic of COVID-19 and with the infection rates on the rise, face masks have become a necessity in public spaces to prevent further spread of the virus. Therefore, it is necessary to develop a system which enforces the need to wear a facemask in public places and identify the ones who are without it. This paper describes a face mask detection and face recognition system. Masked faces are detected using Convolutional Neural Networks and individuals without a face mask are recognized using the LBPH algorithm. The models use OpenCV and TensorFlow and are trained and tested. For doing so, two sets of datasets are used, where the majority of the images are obtained from the Kaggle dataset. 20% of the images are considered for training purposes and other 80% is for testing. The experimental results have confirmed an average face mask detection and face recognition rate of 99%. Accuracy and loss plots for the proposed model are compared with the existing model. Testing for mask detection and facial recognition of masked and unmasked faces are done elaborately.