Real-Time Face Mask Detection Using MobileNetV2 Classifier

Abstract

Since the flare-up of the COVID-19 pandemic, generous advancement has been made in the field of PC vision. The coronavirus COVID-19 pandemic is producing a worldwide health crisis, according to the World Health Organization, and the most effective safety approach is to wear a face mask in communal locations (WHO). Several methods and strategies have been used to develop face detection models. The suggested method in this research employs deep learning, TensorFlow, Keras, and OpenCV to detect face masks. Since it is somewhat asset proficient to introduce, this model might be utilized for wellbeing contemplations. The SSDMNV2 approach uses Single-Shot Multi-box Detector as a face detector and MobilenetV2 architecture as a framework for the classifier. MobileNets are built on a simplified design that builds low-weight deep neural networks using depth-wise separable convolutions. Using a face mask detection dataset, a real-time face mask identification from a live stream using OpenCV is accomplished. Our objective is to use computer vision and deep learning to determine whether or not the individual in the picture/video transfer is wearing a face veil.KeywordsSingle-shot multi-box detectorData augmentationMobileNetV2