Social Distancing Inspection To Mitigate COVID-19 Using K-Nearest Neighbour

The beauty of Indian culture is to celebrate happiness as well as sorrows in gatherings of relatives, friends, and well-wishers. But with the advent of the COVID-19 pandemic, the world order has changed and social distancing has become inevitable and needs to be observed strictly to save mankind. Keeping in view the transmission of the COVID-19 virus through contacts and non-availability of medicines and vaccines, social distancing is one of the key solutions that is way out to contain this spread of the virus. As the name states, social distancing suggests that people should physically maintain distance between themselves which will reduce the chance of close contact and automatically leads to a decrease in the spread of infectious diseases like the Corona Virus and many other viruses which spread through close contact. Social distancing is possibly the most effective non-pharmaceutical way to prevent the spread of disease. The main objective of social distancing is if people are not close together, they are incapable of spreading germs. This mitigation measure is used to decrease transmission, by suspending and decreasing the size of the epidemic peak and spreading cases over a lengthier period to relieve the burden on the healthcare organization. In this paper, a model is recommended where the total number of people present in the frame is detected using the YOLO object detection algorithm, and distance between each individual is measured Using K-Nearest Neighbour. If the distance between any two individuals is less than 6 feet or 2 meters then a red bounding box pops around them indicating that they are violating the rule of social distancing. This model is implemented on Raspberry Pi with buzzer system for alert.