WhatTheFlora!

WhatTheFlora is a mobile app developed to identify and log information about wild edible plants. With each plant, you unlock more of your FloraDex and further your journey to become the Flora Master.

Comment

Inspiration 🌱

Our Inspiration is derived from an anecdote from one of our members. When Samantha was on a hike, she encountered a pecuiliar sight: a vibrant orange flower she had never seen before. Initially, Samatha attempted to google the flower, but she was overwhelmed with the numerous and large variety of flora, and was never able to discern what the flower was. Combining this incident with both the theme of exploration and the team's love for popculture and collection based games, we were inspired to develop WhatTheFlora.

What it does 🍀

WhatTheFlora is a mobile app which allows a user to take a picture of any edible flora they encounter, and then using its image recognition will let the user know the identify of the plant as well as various information about the flora. Once a flora is identified, it will be logged for the user, effectively saving it in their FloraDex.

How we built it 🌳

WhatTheFlora was built by separating the front end from the back end. The front end consists of React-Native, Expo Go, and JavaScript to develop the mobile app. The back end consists of Flask which was used to connect the frontend to the backend, alongside Pytorch which was used to develop a machine learning algorithm for image recognition and classification. A plants API was then used to connect information with each plant, along with MySQL for a database of available plants recognized by the machine learning model.

Challenges we ran into 🌵

A challenge we ran into while developing our machine learning model was not including a save function during the training loop of our model. As a result we had to retrain our data, ensuring to save each epoch. A part from that, the main challenges we ran into were related to connecting the front end to the back end. Since we used Python for the back end and JavaScript for the front end, we had to learn some new skills to figure out how to get everything to work together.

Accomplishments that we're proud of 🌴

We are proud of all the new technology we were able to learn in a short period of time. Going into this hackathon we knew very little about machine learning, and majority of our backend experience was with Node.js. We challenged ourselves to go outside of our comfort zones and learn something new, and we are proud that we were able to complete our project with its intended purpose.

What we learned 🏞️

We learned many new technical skills including the use of Pytorch to develop and train an image recognition model, the use of Flask for a backend framework, and the use of React Native to build a mobile web application. We also improved our skills of combining frontend with backend, especially since we were using new frameworks we were previously unfamiliar with.

What's next for WhatTheFlora! 🧩

  • Develop a log in system, so users can log in from multiple devices and have their progress saved
  • Incorporate a 'Share with Friends' features, where users can add friends on the app and they can share the flora they've found with each other
Share this project:

Updates