Mask Wearing Classification using CNN
Maharani Devira Pramita, Budi Kurniawan, Nugraha Priya Utama · 2020
Indonesian official team for handling COVID-19 stated that from total 358,659 tested people, 42,762 of them are confirmed to be positive COVID-19, approximately 11.7% from total tested people and continues to increase every day. Therefore, wearing a mask becomes a new habit to adapt new normal life, in order to prevent COVID-19 virus spread in Indonesia. It is important for everyone to wear a mask especially in public places with high intense of human interaction. Public area managers need to monitor and ensure everyone is wearing mask properly. To do a better public places monitoring, this work proposes a deep-learning based image classification model to detect whether a person is wearing a mask or not wearing a mask in real-time condition. Proposed architecture is using Convolutional Neural Network (CNN) to learn from annotated private dataset to match the Indonesian characteristics. Result gives 97% accuracy for classifying face.