Inspiration
Long night drives are neither easy nor safe. Having faced this problem personally and failed to find reliable solution, we decided to tackle it by ourselves!
What it does
DriveAlive automatically identifies moments of falling asleep and wakes you up with a sharp sound alarm. What's more, it continuously measures your drowsiness level and recommends to take rest when you're tired.
How we built it
We used face recognition technique to detect any periods when user's eyes are closed for more than 0.5 second. Also based on frequency and latency of blinking we make a recommendation whether driver need to have some rest.
Challenges we ran into
The biggest challenge was to make the application work at night in full darkness when there is no light and phone camera can't detect driver's eyes.
Accomplishments that we're proud of
We succeeded in solving the challenge of recognizing the face in full darkness with building a self-made flash-light with infra-red diodes (IR LEDs) whose light spectrum is invisible for human eyes but can be seen by smartphone.
What we learned
The problem that causes deaths of thousands of people every year can be solved with a usual smartphone's camera and doesn't not require buying a special equipment. The safety of driver and passengers can be significantly improved only with a use of one mobile app.
What's next for DriveAlive
To improve app's reliability it would be reasonable to add such features as Road Marking and Lane Detection that are already available for premium-class cars but practically can be done for any customer having a smartphone.
Built With
- android
- android-studio
- face-detection
- google-vision-api
- java
- machine-learning

Log in or sign up for Devpost to join the conversation.