The Research on Social Distance Detection on the Complex Environment of Multi-Pedestrians

To keep a safe social distance plays an important role in the prevention of high-risk diseases. Aiming at the outbreak of COVID-19, in order to regulate the social distance between pedestrians and reduce the risk of COVID-19 spreading among pedestrians, a multi-pedestrians distance measurement method based on monocular vision is reasonably proposed to realize the measurement of the distance between multiple pedestrians under the monitoring perspective. The pedestrian detection model is used by that method to capture the multi-pedestrians target under the monitoring perspective, and the monocular distance measurement principle is also used to achieve the distance measurement between the multi-pedestrians. Through analyzing these distances, the social distance between pedestrians can be regulated. The experimental results show that this method can efficiently and quickly detect people who do not meet the social distance norms.