Plant Sight is an app that utilizes Machine Learning to detect species of trees and add them to your garden. Can also recognize and report invasive species to your local park authorities.
Our Purpose: We want to heighten interest in the environment around us through fun and interactive educational content. We also want to create a sense of connection between our surrounding nature and humanity.
Our Grand Vision:
- Interactive mobile app for people with or without plant knowledge
- Scan a plant, and then get information on it!
- If invasive, tag it for us to send its location to park authorities
- Grow your garden!
- Each plant you snap is added to your virtual garden
- Look through all your plants at anytime
- View a map that shows the locations of where you found your plants
Our Challenges:
- We had difficulties finding a plant dataset
- Trouble with finding a small version
- Configuring image importation to include labels
- Import through buckets
- Documentation involving different versions of Google Cloud Platform’s AutoML
- Improper placement of dataset and csv file (wrong bucket)
- Could not train model due to ‘internal’ errors
Link to App Demo: https://youtu.be/CVK81A2MGPU
Built With
- auto-ml
- google-cloud
- python
- tflite
Log in or sign up for Devpost to join the conversation.