Plant Guide

What it does

Sometimes we don't know how healthy our house plants are! We built Plant Guide to help you take care of all of your beloved plants and figure out what you need to do to make sure that your plant is as happy as it can be. Users can upload images of houseplants and our website runs the photo through a machine learning model we trained which will determine if the plant is healthy or not. If their plant is not healthy, users are prompted with suggestions of next steps to help them care for the plant.

How we built it

We used Webflow to figure out the UI. We found a dataset of 900 images of healthy and wilted house plants on Kaggle and trained a convolutional neural network that can predict whether the plant is healthy or not. Then we exported the model to Replit and used Python, HTML, CSS, and Javascript to build the website allowing a user to upload an image to be run through the model.

Challenges we ran into

It was challenging to integrate the model into our frontend site and figure out how to get the model's output to trigger actions in the frontend.

Accomplishments that we're proud of

We are so proud that we were able to build a computer vision model and integrate it into our website so that people can interact with it.

What we learned

We learned how to integrate a computer vision model into front-end code!

What's next for Plant Guide

We'd like to add a library feature so users can input what houseplants they have and get a suggested watering schedule for all their plants together. In the future, it would be useful to add user profiles and authentication so users can return to the site and pick up where they left off with their plant library. Additionally, we see the possibility for a vision model like this to be used to identify the condition of larger farm lands in the future where the health of the crops is important.

Built With

• css
• html
• javascript
• kaggle
• python
• replit