Mask Detection and Facial Recognition System Using a Convolutional Neural Network Algorithm

During the current COVID-19 pandemic, wearing a mask is mandatory for everyone when engaging in various activities to prevent the spread of the COVID-19 virus. So far, the detection of mask use has been done manually through observation by officers. This method has several limitations, namely that it cannot be performed every time and in every place. Based on these problems, the researchers designed a Mask Detection System and Face Recognition Using Convolutional Neural Network. This system is designed for the Department of Information and Computer Technology at the Lhokseumawe State Polytechnic. The system input is in the form of a real-time video camera that can detect students both wearing and not wearing masks. Every student who does not wear a mask will be recognized by their face. The system will send a notification to the officer if any students are not wearing masks. The system utilizes a mask dataset sourced from the Kaggle website, and for facial recognition, it employs facial data from 4C IT class students. The results obtained in the making of the CNN model that were tested resulted in 99% accuracy, and real-time testing resulted in an overall accuracy rate of 75%, mask detection accuracy of 83%, face detection accuracy of 71%.