Inspiration

We all know how bad COVID has been. There has been over 114 million cases, and over 2.5 million deaths worldwide. even worse is that we could have prevented most of these deaths by properly social distancing and wearing a mask. With the 24 hours I had to create, I wanted to build something fun but also wanted to build something where I learned something new. Hence, I created this project.

What it does

uses openCV and machine learning to detect social distancing. It uses object detection to detect if something is a person or not. If it is it gets the centroid of the person and uses brute force to see if the centroid of each person compared to another person is less than or greater than 6 feetl

How we built it

OpenCV, python

Challenges we ran into

-- first time using openCV in a long time --first time making a complex OpenCV project -- had trouble upgrading my computer to the latest version for the OpenCV -- never had used Deep Neural Networks before

Accomplishments that we're proud of

-- first time using a DNN -- 90%+ accuracy -- Had fun while making the project

What we learned

-- Deep Neural networks -- Open CV

What's next for SocialDistanceAI

-- Try scaling this idea. -- Improve the camera so it has an alarm system when one is not social distancing.

Built With

Share this project:

Updates