Face Mask Detection Using Deep Learning

Abstract

The COVID-19 epidemic has highlighted the necessity of wearing face masks to help prevent the virus from spreading. Automated face mask detection systems have drawn a lot of attention because they can help enforce mask-wearing rules. For precise and immediate face mask identification, this project suggests using the MobileNetV2 algorithm, a cutting-edge deep learning architecture. The project starts by gathering a sizable collection of labeled photos of people wearing and not wearing face masks. These photos have been carefully chosen to represent a range of perspectives, lighting circumstances, and mask types. The dataset's variability is increased by using data augmentation techniques including rotation, scaling, and flipping, which improve the model's generalization skills. The MobileNetV2 architecture is then developed on top of the face mask detection model. Real-time applications notably benefit from MobileNetV2's effectiveness and flexibility for deployment on low-resource devices. The pre-trained weights of MobileNetV2, which were trained on a considerable amount of image classification jobs, are used as a starting point in order to benefit from the network's feature extraction capabilities.