Face Mask Detector using Deep Transfer Learning and Fine-Tuning
Shashank Sai Palani, Mayank Dev, Giridhar Mogili, Devanjali Relan, Rajiv Dey · International Conference on Computing for Sustainable Global Development · 2021
The world Health Organization declared a pandemic called the coronavirus. To avoid the spread of coronavirus disease (COVID-19), it released an advisory of some preventive measures to be taken care of while dealing at public places and health-care centers that include wearing masks, social distancing, monitoring health, and using disinfectants. In this paper, we have taken one of the measures used to prevent COVID-19 spread and aimed to develop a deep learning model to categorize people with or without a mask at public places such as schools, colleges, and corporates. In this paper we developed algorithm using concepts of deep transfer learning and fine-tuning. The system developed on the MobileNetv2 base model, the head of it was replaced by the custom face mask detection algorithm and enabled the training of face masked and non-face masked images. The testing results have shown an accuracy of 98% on both categories with mask and without mask.