Social Distancing Violation Detection Mechanism Using Opencv

EasyChair Preprint no. 10143

Abstract

To counter the pandemic of Novel CoronaVirus one of the most effective measures was Social Distancing. However, a lack of spatial awareness may unintentionally transgress this new need. In light of this, we suggest a proactive surveillance system to contain the spread of COVID-19. We offer a real-time image-based system that uses deep learning models to identify Social Distancing violations and transmit visual information. The injury probability Social Distancing can therefore approach zero if the pedestrian density stays below a newly defined critical social density threshold, which we then specify. The proposed system is also morally sound. We do not collect information or target specific individuals. There are no administrators working. It can be integrated and used with the cameras installed in the public places to monitor the situation and condition of social distancing.

Keyphrases: linear regression, pedestrian detection, social distance

@Booklet{EasyChair:10143,
  author = {Samarth Sachan and Shreedhar Sharma and Utkarsh Chaudhary and Mukesh Raj},
  title = {Social Distancing Violation Detection Mechanism Using Opencv},
  howpublished = {EasyChair Preprint no. 10143},

  year = {EasyChair, 2023}}