Face Mask and Social Distancing Detection in Real Time

Face Mask and Social Distancing Detection in Real Time

Madhumita Choudhury, Durba Paul, Anal Acharya, Nisha Banerjee, Debabrata Datta
Copyright: © 2023 |Pages: 24
DOI: 10.4018/978-1-6684-8306-0.ch011
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

With the recent outbreak and rapid transmission of the COVID-19 pandemic, the need for the people to follow social distancing and wear masks in public is only increasing. So, the main objective of this chapter is to build a machine learning model based on TensorFlow object detection API and YOLO Objection Detection that will determine a green and red rectangle around the face if the person detected in the camera wears or does not wear a mask, along with an email alert being sent to the authority in charge informing about a person's violation of face mask policy and will return a green or red bounding box accordingly if social distancing is maintained between two people and at the same time alert others by a beep alarm. The accuracy of the model is nearly 97% so it can be used by governments to alert people if the situation turns serious.
Chapter Preview
Top

This section previews some of the related research works on Object Detection, Object Recognition, Face detection and identification based on Artificial Intelligence, Convolution Neural network, YOLO Object detection algorithm, TensorFlow Object Detection API, OpenCV package, etc., and implementation of these models in real world applications. Several works have been done successfully in this domain. Many of these significant works have been thoroughly examined to gain knowledge in this field and build the proposed model for our research paper:

Zhong-Qiu Zhao, Peng Zheng, Shou-tao Xu, and Xindong Wu (2019) attempted to provide deep learning-based object detection frameworks in a research paper and compared various object detection methods on three benchmark datasets, including PASCAL VOC 2007 (n.d.), PASCAL VOC 2012 (n.d.) and Microsoft COCO. This paper also focussed on the modifications that can be made to typical generic object detection architectures and also surveyed tasks like face detection, pedestrian deletion etc.

The main research work of the article put forward by Sheng Ding and Kun Zhao (2018) was to collect a small data set of daily objects and using the TensorFlow framework, build different models of object detection, and use this data set for training model and finally effect of the model was improved by fine-tuning the model parameters.

Complete Chapter List

Search this Book:
Reset