LeafX

Inspiration

Throughout California, we noticed how there were many plants that had different colors and unknown substances on the leaves. We wanted to figure out what was going on inside of the plants, specifically the medical health for these plants. Plants are important within society, as they provide us oxygen, and every plant plays a big role in dealing with climate change. We were inspired by non-profit organizations such as Team Trees that help combat climate change and reforestation.

What it does

This app allows its users to upload images of their plants, and our machine learning model detects if it is infectious or healthy. If it is diseased, our app will present information regarding the disease.

How we built it

We used React Native for all the front-end engineering, and we used Tensorflow.js for all the machine learning.

Challenges we ran into

The challenge we faced was fully integrating Tensorflow.js into our project. We spent hours during the night trying to figure out why our app was bugging out when we wanted to run a prediction model. However, after hours of searching Youtube and Stack Overflow we were able to conquer the situation.

Accomplishments that we're proud of

Our greatest accomplishment was our success in implementing Tensorflow.js into our project. It was our greatest challenge yet since neither of us were experienced. We spent a total of 12 hours just trying to train and integrate our model and it required us to start over from scratch multiple times.

What we learned

We learned how to use Tensorflow.JS to react-native, while also the fundamental methods that scientists use to determine a plant's health.

What's next for LeafX

We hope to enhance our machine learning model by implementing a more accurate model by training with more classes (more diseases) and more image samples, hence yielding a more accurate model.

Built With

• javascript
• reactnative
• tensorflow.js