In the midst of a pandemic, we decided we needed a better way to determine whether people were wearing masks correctly in order to enhance security and safety.

What it does

The site has a built-in webcam that detects faces and masks. It will tell the user whether the mask is being worn properly or not.

How we built it

We built the site with raw HTML, CSS, and JS creating an appealing style for our users. We used image recognition to detect faces and masks to be able to tell our users whether or not they were wearing their masks correctly.

Challenges we ran into

Running flask and OpenCV was quite difficult on REPL. Along with collaborating on the same portion of the project since it is online and REPL was not functioning well.

Accomplishments that we're proud of

We are proud of the webcam on our site being able to detect faces and masks.

What we learned

Communication and collaboration are important for a successful team. Goal-setting and time management are vital in creating a wonderful project like ours.

What's next for Mask Verified

We hope to create an app in the future that would have the same features but it would be more applicable to daily life.

Project details

Our project locates eyes in a camera's frame and determines the size of the head based on the distance between the centers of the eyes. The area of the frame with the head is then turned into a numpy array and resized to dimensions that are inputted into our AI model. We used tensorflow to train a model on images of faces with and without masks, and it is 97% accurate with < .1 loss. Finally, it tells the user on the website to put on a mask if they are not wearing one or congratulates them if they are wearing one. We used flask as our interface.

How this applies to the real world

Safety is a top priority, especially since the COVID-19 virus (and its variants) is still spreading. Keeping masks on lowers the chance of the virus spreading. This project can check people who are entering stores if they are wearing a mask or not and give them feedback on their decisions to better society and save lives.

Built With

Share this project: