Face Mask Detection Using Deep Learning Techniques

Abstract

The SARSCoV2 that causes coronavirus disease 2019 (COVID-19) was transmitted in Wuhan, China, in December 2019. Later, the virus started to spread from person to person. Face masks are a type of personal protective equipment that can assist to prevent the spread of respiratory illnesses, viruses, and bacteria. Face masks are utilized to prevent the spread of COVID-19 infection. Hence, face mask recognition is still a challenging problem in computer vision. To propose a solution for face mask recognition the proposed system uses a combination of Convolutional Neural Network and specific image pre-processing steps.  It describes the innovative solution that provides efficient face mask detection and deep learning with convolutional neural networks (CNNs) comparing various architectures and using LeNet to achieve great success in the classification of face masks, especially for human safety using Anaconda and Jupyter. A variety of neuron-wise and layer-wise visualization methods were applied using a CNN, trained with publicly available datasets and from a given image dataset. So, it’s observed that neural networks can capture the colours and textures of lesions specific to the respective face mask, which resembles human decision-making.