Inspiration
Water and sanitation is recognized as a basic human right, however much drinking water around the globe is not safe to drink. Aquabac aims to solve this problem, by letting a user know whether drinking water is safe or not in a quick and convenient manner.
What it does
Aquabac is a machine learning system that detects bacteria in water targeted towards home use. Aquabac is able to detect more than 33 different types of common bacteria at 99% accuracy. Our system takes in images of water samples with bacteria. Using advanced algorithms, Aquabac detects whether the water sample has bacteria in it, identifies the species, and then tells the user information about it and any important things to know about the species. Aquabac is user friendly, requiring only pictures in order to output results.
How we built it
- Google Cloud for machine learning
- Domain.com for web hosting
Challenges we ran into
- Internet issues
- Finding valid data
- Running out of cloud credits
Accomplishments that we're proud of
- Creating a machine learning system with more than 99% accuracy
- Identifying the most prolific bacteria strains
- Implementing a decent UI in a relatively short timespan
What we learned
- Utilizing Google Cloud system for machine learning
- About resource needs around the globe
- Dashboard creation
- UX/UI creation
- Cloud connection
- Databasing
What's next for Aquabac
- Implementation of Aquabac on a larger scale
- 24/7 system that automatically sends images from the site back to Aquabac for results
Log in or sign up for Devpost to join the conversation.