Deep Learning Model based Face Mask Detection for Automated Mandation
Pratham Lokhande, Shivangi Surati, Himani Trivedi, Bela Shrimali · 2023
SARS COVID19- virus has been a serious threat to human life, the best way to protect oneself is by wearing a face mask. As per World Health organization (WHO), wearing a mask can cut down the risk of infection by 85%. This study sought to offer an automated solution using several state-of-art Deep Convolution Neural Network (CNN) face recognition models to classify an individual’s face is wearing a mask or not. Determining the same and regulating the same is a difficult task that involves great human efforts. Nine various CNN models are trained, thereafter tested and found that the architecture of NasNet outperforms the other eight state-of-the-art models with respect to accuracy, on both, Real-World Masked Face Dataset (RMFD) and Simulated Masked Face Dataset (SMFD). The proposed approach achieved 100% testing accuracy on SMFD and 99.872% testing accuracy on RMFD.