Based On Segmented Video Image Processing, Person Detection for Social Distancing and Safety Violation Alert

ELECTRONICS & COMMUNICATION ENGG.

Both humans and animals are harmed by the broad family of coronavirus viruses. The coronavirus identified as Covid-19 initially made its way to Wuhan, China, in December 2019. We can alter the pixels to either improve the image's quality or get more relevant data out of them. Other than measuring social distance, the device can also identify persons in restricted zones and utilise that information to provide warnings. The program's efficacy for both goals has been tested through some analysis. According to the results, the distance tracking system tested indoors in a controlled setting and achieved 100% accuracy, whereas testing outside required tougher input footage and the system only managed 56.5% to 68% accuracy. Contrarily, it was discovered that the safety violation alert feature based on segmented ROI had obtained greater accuracy, ranging from 95.8% to 100% for all of the examined input films.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.