AUTOMATIC SOCIAL DISTANCING AND FACE MASK VIOLATION DETECTION IN PANDEMIC SITUATION USING YOLO & TENSOR FLOW

COMPUTER SCIENCE & ENGINEERING

Virus epidemics might be moderated if individuals consent to orders to remain at home and stay away from strangers in the open. All thingd considered, there is general wellneing interest in social distancing consistence. The accessible proff on distancing rehearses out in the open space is restricted, be that as it may, by the absence of observational information. Consistence with 1.5 meter distance orders is brief and harmonizes with the quantity of individuals in the city and with consistence to remain at-home orders. Expected ramifications of these discoveries are that keep-distance mandates may work best in mix with stay-at-home orders and spot explicit group control methodologies, and that the quantity of individuals in the city and local area wide portability as caught with cell information offer effectively quantifiable intermediaries for the degree to which individuals stay away from others at explicit occasions and areas. Here the system uses Yolov3 and Tensorflow for implementation. Yolov3 is precompiled library through which object classification can be done in a very effective manner as compare to the other conventional techniques. Tensorflow helps Yolo to classifies the objects with high preciseness. System is able to detect pedestrian with high level of accuracy and it has been designed to detect the social distancing rule violation as per their physical distance which has been measured as per their appearance. System achieved 93.84% of accuracy

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