-
Home Screen
-
Plant Recognition Screen
Credentials for testing: email: testuser@example.com password: SfGh7w9=5y3xj8)P
Inspiration
We were inspired by the random plants we receive as house warming gifts that we have no clue what the name of the plant was or even how to take care of it.
What it does
PLID (Plant Identifier) is an AI powered plant identification server that lets users upload images of their house plants to identify the species and how to take care of the plant.
How we built it
We build PLID as a full stack web app. The backend is comprised of mainly python code using flask for our web routing and Keras for training our neural network. We also used MySQL as our database and auth0 to verify logins. The frontend was built using React, Typescript, and Material UI so that we could ship a good looking product quickly.
Challenges we ran into
We can into a major overfitting issue because of our small data set, we offset this challenge by making many permutations of our dataset to simulate different lighting and contrast conditions.
Accomplishments that we're proud of
We're proud of the accuracy of the model especially with the limited time and training.
What we learned
We learned how to quickly ship a minimum viable product, how to use auth0, and how to preform image classification with small datasets.
What's next for PLID
We plan to use this as a learning experience to further our careers.
Log in or sign up for Devpost to join the conversation.