We know Prevention is better than cure, so the happy driver not only warns you but saves others life. We have chosen healthcare as the track because there is nothing more important than saving life. “Safety isn’t expensive it’s priceless” so we Team Aliferous come up with a solution “ Happy driver” to save your life free of cost.
What it does
The Happy driver is an automated eye-tracking web app to detect the drowsiness of drivers. The detector monitors your eye movement and plays an alarm to remind you that you need a break, and saves your life accordingly. Sleep-deprived drivers remain responsible for about 40% of road accidents. This can be one stop solution for 40% of people in the whole world. According to stats 1.35 million people dies each year as a result of road accidents, using happy drier we can save potentially all those lives.
How I built it
We have used dlib library for face landmarks detection and then calculated eye aspect ratio(EAR). EAR can be used to decide if a eye is open or closed. We have used opencv and base64 for image processing and Django has been used in the backend to make the web application.
Challenges I ran into
As we did not receive SSL certificate from AWS, the camera was not opening after deployment. So we were not able to deploy the application and needed to run it on local server. We wanted to make a blink counter, which would let users know how much rest they need, but the alarm and blink counter are not working simultaneously. We wanted to connect it with google map API but were not able to manage time. We even tried to add some svg animations but weren't able to do as well. Not the least , we also tried to build our own API but were not to be do it.
Accomplishments that I'm proud of
Never connected a machine learning model with Django backend we were able to connect them. Accidents due to drowsiness is real world problem and Happy driver is a one stop solution for it. The machine learning model does not consume much resources and can be deployed with minimum resources and has an efficiency of around 90%.
What I learned
Time was the biggest barrier as this was our first hackathon we learned about time management. While developing this project we also learned how to use machine learning model with Django as backend.
What's next for Happy Driver
Undoubtedly, everyone uses maps while driving interacting happy driver as a plugin will full fill our need. Happy driver may be used as a plugin in software like Google Maps, so that the application will start functioning whenever the user opens Maps and click on start.