CNN-based Mask Detection System Using OpenCV and MobileNetV2
Gregor Christa, J Jesica, K. Anisha, K. Martin Sagayam
Abstract
this paper establishes a `Safety system for mask detection during this COVID-19 pandemic'. Face mask detection has seen an overwhelming growth in the realm of Computer vision and deep learning, since the unprecedented COVID-19 global pandemic that has mandated wearing masks in public places. To tackle the situation, machine learning engineers have come up with several algorithms and techniques to identify unmasked individuals using various mask detection models. The proposed approach in this paper adopts frameworks of deep learning, TensorFlow, Keras, and OpenCV libraries to detect face masks in real time. The trained MobileNet model, presented in this paper, yielded an accuracy score of 0.99 and an F1 score of 0.99 in the training data. This user-friendly model can be incorporated with several existing technologies such as face detection, biometric authentication and facial expression detection for further advancements in the future.