COVID-19 Social Distancing Detection and Email Violation Mechanisms

Abstract

The technique of flattening the curve for coronavirus-infected cases is challenging in addressing the worldwide ongoing rampant novel COVID-19 pandemic crisis unless citizens take steps to halt the virus’s spread. Maintaining a safe space between individuals around us in public is one of the most important behaviors. Deep learning algorithms have been used in the proposed work to mitigate the spreading of the coronavirus utilizing social distance detection. This proposed work analyzes a pre-recorded video feed of walking pedestrians to alert people to maintain a safe distance. The goal is achieved using YOLOv3 and YOLOv4 for object detection in the video frame used as input. Furthermore, an email-based alert mechanism is also implemented if the number of violations exceeds the defined limit. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.