Only 9% of plastic waste is recycled in Canada and similarly only 1/10th of solid waste is recycled in the United States.

What it does

Using machine learning and computer vision we are able to accurately classify waste into 7 different categories (cardboard, paper, metal, glass, plastic, organic, and trash). The machine learning algorithm I trained was able to achieve 90% correct classification of waste.

How I built it

Cross Platform Mobile App

The mobile app was built using React Native in the front end and using a Flask server for the backend.

Machine Learning Algorithm

This algorithm was built using the TensorFlow and Keras frameworks.

Accomplishments that I'm proud of

I am proud of the accuracy that I was able to achieve in the machine learning algorithm. I am also proud of the smooth user interface that I was able to develop in React Native.

What I learned

  • Multiclass image classification.

  • Computer vision

  • React Native

What's next for ReGreen

Built With

Share this project: