We wanted to prevent the accidents and deaths falling asleep at critical times causes.
What it does
Notifies you with a large sound when you are asleep on webcam so that you can go back to working.
How we built it
We used Flask to run Python code on the web. We trained a model with Keras and saved it as an h5 file. We used OpenCv to locate the person's eyes and then fed them through this neural network to determine if they were asleep.
Challenges we ran into
It was hard to get constant pictures from the video stream and integrate Flask with the website.
Accomplishments that we're proud of
We are proud that we got it all to work and that we made something that is useful in the real world.
What we learned
What's next for StayUp
Android and IOS Application More training (layers and epochs), Further testing, Adding optional functionality such as a math problem to verify that the user is truly awake