I was very excited about the outdoor theme for BackyardHacks, as I've worked in agriculture for many of my summer jobs. I wanted to create a feeling of wonder and curiosity about exploring the world around us.

What it does
Florafy uses your phone's camera to classify various plants.

How I built it
So, my dataset and model are both stored in Google Cloud. I trained my dataset using Firebase's AutoML, as well as used their image analyzer in my code. The app itself is written in Kotlin, with the front-end in XML, using Android Studio as the IDE.

Challenges I ran into
The biggest challenge was finding data to train. I combined a couple datasets from Kaggle with some labels I made myself.

Accomplishments that I'm proud of
This is actually my first time venturing into the world of machine learning and computer vision, though it's something I've always been interested in! So I'm proud that I was able to create this app with many new concepts in only 48 hours!

What I learned
I learned a ton about machine learning and computer vision, and I'm looking forward to exploring these tools more in the future!

What's next for Florafy
For Florafy, the biggest task is training a new model with a bigger dataset so it can recognize almost any plant!

Share this project: