Due to the COVID pandemic that has been going on since 2020, there has been many new issues
arising that could be solved by technology and software. One of them is a lot of new rules being put
into place to protect people from getting infected with the illness and naturally a lot of people refuse
or don’t know how to follow the rules properly. Current solutions on how to enforce these rules is to
have a person standing there trying to catch as many people that break the rules as possible. This
solution is feasible, but by using a software to identify the breaches, the identification will be more
accurate and will catch more of the offenders than when done by a human, especially in large and
crowded areas. This will also take workload off essential workers, as during the pandemic there isn’t
enough of them to do all the tasks that were added onto the ones they already had to do. And even
when this pandemic is over this software could be used in workplaces, where masks mandatory like
hospitals and laboratories, or if another pandemic occurs this software can be deployed straight
away instead of waiting for it to be developed.
The solution for my program lies in the field of artificial intelligence and computer vision and after
further research combined with my supervisor’s recommendation, I decided to use one of the object
detection algorithms. I did a lot of research on these algorithms to find, which one will be the best to
use in this case and in the end, I decided to use the YOLOv4 algorithm and for my coding
environment I went with Google Colab. Besides developing my product, I also researched ways to
improve the accuracy of the YOLOv4 algorithm.