Monitoring Social Distancing for Industries and in Public Areas Using Machine Learning

ABSTRACT

The deadly coronavirus disease 2019 (COVID-19) has been spreading vigorously and has led to a global crisis, with its spread to more than 220 countries and territories. About 153, 504, 608 confirmed cases of the coronavirus COVID-19 that originated from Wuhan, China, and a death toll of 3, 216, 577 deaths as on 3 May 2021. At the time of writing, India is the worst affected country by COVID-19 and the death ratio is increasing day by day. To date we have more than four vaccines available, but social distancing has been identified as the best way to overcome and fight against this disease. In order to ensure social distancing protocols in overcrowded places and workplace, this tool can monitor whether or not people are ensuring a safe distancing protocol from each other by analyzing real-time video streams with the help of a constant camera feed. To keep track of people in various workplaces, factories, and shops we can use this tool to their security camera systems and can monitor whether or not people are keeping a secured distance from one another. This chapter proposes a Machine Learning and Python-based framework for monitoring social distancing using surveillance video with the help of a camera. In this proposed framework, we are utilizing the YOLO v3, an object detection model for separating the foreground details from the background details and OpenCV for tracking the humans by using the bounded boxes and assigning IDs to them. © 2022 selection and editorial matter, Abhinav Goel, Anand Chauhan and A.K. Malik.