Inspiration Preventing driving accidents caused by sleep deprivation
What it does Plays a loud noise when Watson determines that your pictures shows that you are too drowsy
How I built it We used the raspberry pi as the camera to take a picture every 10 minutes. The picture is sent to Watson Facial Recognition API, which runs the facial data against a group of pictures showing sleepy and attentive people If the driver is catagorized as sleepy, the the raspberry pi receives a signal to play a beep.
Challenges I ran into: Interfacing between the Watson API and the raspberry pi. The Watson API protocol was very novel and proved to be challenging
Accomplishments that I'm proud of include running script in python to automate the picture capturing process. Thinking of a novel and useful way to integrate Watson Facial Recognition into driving.
What I learned We learned how to code in python on a raspberry pi. We learned about Watson's new API's, specifically the Facial Recognition algorithm. We learned about the differences in language between Watson and raspberry pi.
What's next for Attentive Driver Integration with an accelerometer so we can make sure the device is only on when the car is moving.