Scallywag Spotter

Title Page of our Website
Promotion of our product
More promotion of our product
Cognitive Test
The creators behind the product
Our 3D printed breathalyzer phone case (without the sensor in it)

Inspiration

Drunk Driving is very dangerous and kills many people each year. So we made a project that detects intoxication to prevent accidents like these and fix existing flaws in existing products,

What it does

Scallywag Spotter is an advanced dynamic detection software that leverages hardware for precise intoxication detection. Our system employs the use of an Arduino to accurately distinguish alcohol levels in your breath. To further enhance accuracy, our comprehensive IQ quiz is administered both prior to and after the detection process, enabling us to thoroughly analyze and identify any noticeable differences. With Scallywag Spotter, you can trust in its effectiveness and precision for accurate intoxication detection.

How we built it

We used an Arduino and MQ3 sensor to create a breathalyzer for an extra layer of drunk detection. We also 3D printed a phone case design to hold the breathalyzer. However, breathalyzers aren't always reliable: this is demonstrated in our video by setting off the breathalyzer with mouthwash. So, to an add an extra layer of security, we used an IQ quiz for successful confirmation. All the data was put into a user-friendly website.

Challenges we ran into

Some challenges we ran into was connecting the Arduino serial data and implementing a ML/CV model. We spent the ENTIRE first day attempting to create a model that would detect if you were drunk simply by looking at your face. However, we had to scrap the model as we ran out of time.

Accomplishments that we're proud of

We are proud of the hardware connection and the formal website we developed.

What we learned

We learned how to create websites using react and connecting arduino serial data into the website. We also learned that machine learning is very complicated...

What's next for Scallywag Spotter

We're planning to incorporate machine learning into the detection of drunk faces. We also plan to use a computer-vision framework to use the detection.