Inspiration
We're inspired by the need to provide fresh produce and reduce food waste.
What it does
Our program identifies a produce product, determines its condition, surrounding temperature, and surrounding humidity, and displays the information on an admin console.
How we built it
Utilizing an uno q arduino, we use a webcam to provide a video stream. Then, we custom trained a computer vision model to distinguish between fresh and rotten fruits. Attaching a temperature/humidity sensor gives us the live data, which we then track and store in an sql database on the arduino. The arduino handles all the backend, and then hosts a web-server with the front-end information, communicating freshness, temperature, humidity, and the type of fruit all to the client.
Challenges we ran into
We initially struggled with training the cv and setting up arduino peripherals. We also struggled with only having one arduino because it limited its use to one person. This created a parallel workflow.
Accomplishments that we're proud of
We're proud of the front end side of our development and the cv model.
What we learned
We learned how to train a cv to identify varying levels of freshness in produce.
What's next for Apple Core
We can increase the strength of the cv to be more accurate.