Inspiration
Over 100,000 car accidents and 15,000 fatalities are caused by people who aren't paying attention to the road. One of the biggest causes of these accidents is a driver's overwhelming stress, people who are drowsy, and people in semi-unconscious states. Therefore, we wanted to create an app to help avoid more accidents by creating DrowsiDetect
What it does
DrowsiDetect is a web application that checks if the driver is feeling drowsy while driving by taking screenshots to examine their physical condition and then taking speed-oriented tests in order to measure their focus and attention skills at the moment and at random moments of your drive/trip. If the person is not drowsy, then the website will display that it is safe for you to drive, else it will display that it is not safe for you to drive.
How we built it
We built a deep learning model to detect if the person's eyes are awake or asleep. If the AI thinks that the person is asleep, we do another series of tests on the person to check if the AI is correct or not
Challenges we ran into
We were unable to connect the front-end tests to the back-end AI at the end of our time
Accomplishments that we're proud of
We built both tests.
What we learned
That we need to connect front-end to back-end
What's next for DrowsiDetect
To connect front-end to back-end.
Built With
- dlibs
- imutils
- javascript
- json
- machine-learning
- numpy
- pip
- python
- react
- scilearn
- tensorflow
Log in or sign up for Devpost to join the conversation.