A Novel Approach for Face Mask Detection using Tensorflow for Covid-19

Corona Virus Disease (COVID-19) pandemic has become a cause world health crisis. It is a disease that can spread from humans to humans through physical contact with the infected droplets or via airborne. It has been scientifically proven that wearing a face mask is the most effective method against...

Full description

Saved in:
Bibliographic Details
Published in2021 Third International Conference on Inventive Research in Computing Applications (ICIRCA) pp. 1362 - 1366
Main Authors Mishra, Sushil Kumar, Ojha, Nitish Kumar
Format Conference Proceeding
LanguageEnglish
Published IEEE 02.09.2021
Subjects
COVID-19
Data models
Detectors
Electric breakdown
Face recognition
Facemask detection
Mobile_NetV2
Pandemics
Tensor Flow
Training
Online AccessGet full text

Cover

More Information
Summary:Corona Virus Disease (COVID-19) pandemic has become a cause world health crisis. It is a disease that can spread from humans to humans through physical contact with the infected droplets or via airborne. It has been scientifically proven that wearing a face mask is the most effective method against the virus. This paper's aim is to develop a face mask detector which could be used to make mitigation, evaluation, prevention, and action plans against COVID-19 pandemic by the authorities. In this study, the face mask detection is developed based on the image classification method called Mobile_NetV2. The pseudo-steps for making the detector model are accumulating data, pre-processing, breakdown of the data, training the model, and implementation of the model. The proposed model is able to detect the people with or without a face mask with an accuracy of 96.85 percent. The experimental results of the model have been performed on real-time applications. The mask detector is also able to detect the face mask on a moving subject with expected accuracy.
DOI:10.1109/ICIRCA51532.2021.9545019