Abstract:
Human Health is turning into a major issue even in this cutting-edge period of innovation. The
COVID-19 that was declared a pandemic by the World Health Organization (WHO) on March
13, 2020, after the widespread of the disease in Europe, and the drastic number of deaths in Italy
is exceptionally transmissible. The current scenario has shown quite a noticeable impact on
people's mental well-being besides their physical well-being. It has changed their perception
about life, and their priorities regarding daily life routine have also been affected. Such a global
situation can only be controlled with people's consent to behave in a particular manner instructed
by health care providers such as frequent hand washing, using the facemask, avoiding gatherings,
and maintaining permissible distance. But we seem to have no such mechanism to avoid such
situations. So, out project reiterates on creating a device which can help in detecting the amount
of people who are following SOPs regarding wearing of face mask through camera feed. This
can assist in minimizing the damage caused by COVID-19. The device has two modules, one is
the device itself consisting of Nvidia Jetson Nano which will be connected to Camera and a
Display Screen, the second one is the training model mainly Yolo v5. The Camera live feed is
sent to the Nvidia Jetson which will process the video, frames by frames and by using YoloV5
trained models will identify the person wearing masks or not. The authorities can view the
detailed aspect on the Display Screen. The principle preferred standpoint of this framework is
that the client does not require an expensive device to differentiate between Masked and Nonmasked
persons dissimilar to different devices that have been developed before. Furthermore,
utilization of refined segments guarantees exactness and makes it robust.
