Inspiration
I think we're all concerned about microplastics; research shows that we can find these pesky chemicals in almost all parts of our bodies. So we wanted to find an accessible way to mitigate microplastic exposure.
What it does
We have hardware that can transform a phone camera into a makeshift microscope. Then, our website allows you to take pictures of water with your phone and uses an AI model to predict whether the water has microplastics. In order to obtain better results, we shine different colors through the water while taking our photos. It is very cost effective.
How we built it
- We disassembled a laser pointer and used its lens to create a microscope
- We hooked up an LED strip to an Arduino UNO R3
- We use Gemini API to pick the colors to shine through the water
- Our site once again uses Gemini API to predict the concentration of different contaminants in the water
- We have a python backend server to communicate with the Arduino ## Challenges we ran into
- Finding components
- Using a wide range of APIs(i.e. Gemini)
- Interfacing our website with the Arduino
- SPUR wifi network ## Accomplishments that we're proud of
- Being able to actually interface with our Arduino
- Implementing AI to predict contaminants in water ## What we learned
- How to use an Arduino
- Gemini API
- Resourcefulness ;) ## What's next for Plastif.ai
- Use a better AI model
- Use better lighting (Dollarama LED strips might not be the play)
- Adding in different sensors


Log in or sign up for Devpost to join the conversation.