Inspiration
We were inspired first by the excitement shown over crisp, delicious apples. To help preserve those apples, they hype around them, and answer the question provided to us by Agritech, we decided on PlantMD.
What it does
You provide us with either your symptoms, or a picture of your plant, and we provide you with information and solutions.
How I built it
We used an ML model trained on 87,000 leaf images for image recognition, and we provide fields for customizable input in order to determine what disease(s) you are presenting to us. We then check our database of web-scraped info to determine solutions for you, and point to you external products known to be helpful in solving your issue.
Challenges I ran into
Introducing ourselves to new frameworks for development (Fast API, React) and uploading images to google cloud storage to use on google vision AI were some of the most difficult challenges we ran into.
Accomplishments that I'm proud of
We accomplished most of what we set out to do, and our UI is minimalistic and cleaned up from the start!
What I learned
We learned how to incorporate CV and an ML model into a front end application.
What's next for PlantMD
We hope to broaden the knowledge of our model, and increase the data we have for plant diseases to account for more crops!
Built With
- alexnet
- cockroachdb
- computer-vision
- css
- fast-api
- google-app-engine
- google-cloud
- html
- javascript
- machine-learning
- neural-networks
- python
- react

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