EnRoute

• 

  Home Screen

Inspiration

Drunk driving claims far too many innocent lives every year, making it one of the most preventable causes of accidents on the road. Despite strict laws and awareness campaigns, many individuals still make the dangerous decision to drive while under the influence, putting themselves and others at risk. Our inspiration for this project comes from the urgent need to reduce these tragedies and create a tool that helps drivers assess their own sobriety before getting behind the wheel, ultimately preventing accidents and saving lives.## What it does The app performs a horizontal gaze nystagmus test, simulating a field sobriety test commonly used by law enforcement. It tracks subtle, involuntary eye movements to help determine if someone is under the influence of alcohol. Additionally, it includes multiple facial recognition checkpoints throughout the process, ensuring that the same user is completing each step of the test. This prevents users from trying to bypass the system by having someone else perform part of the test for them, ensuring the accuracy and integrity of the results.

How we built it

We utilize a self trained facial recognition model along with google's ML pipeline, mediapipe to run a Horizontal Gaze Nystygmas test. We used openCV for image processing and numpy for all the array calculations to compute the gradients and determine whether nystygmas were present.

Challenges we ran into

One of the biggest challenges we faced was ensuring that the same individual was performing every step of the test, particularly to prevent any cheating or manipulation. This was critical for maintaining the integrity of the sobriety assessment.

Accomplishments that we're proud of

The biggest accomplishment we had was being able to make sure that the correct user was the one that was driving and not just anybody. This ensured that our test could not be doped at any stage and that only a sober driver would ever be in the driving seat.

What's next for EnRoute

We really like the idea and want to pitch it to a few government organizations to see if we can begin the process of having a regulation model like this embedded within all cars. Additionally we plan to pitch the idea to Y combinator and Telora to hopefully accelerate us into the startup pipeline.

Built With

• facerecognition
• flask
• gradient
• matplotlib
• mediapipe
• multithreading
• opencv
• swift
• tensorflow
• uikit