Inspiration

after doing some research we found that 146.1 million tons of Municipal solid waste were landfilled. 18% (26.3 million tons) of plastic and 12% (17.53 million tons) makeup the landfill. These are materials that can easily be recycled greatly reducing carbon emission, which would lead to a cleaner world.

What it does

Recyclable allows users to easily classify the material of an item by using ML model that analyzes and labels the input image. The app also shows user how much Co2 he / she is saving by recycling. The app also has a leaderboard where people can "compete" to recycle.

How we built it

  • Java
  • ML model trained using AutoML Vision with 2,300 datasets.
  • Android Studio
  • XML
  • Firebase

Challenges we ran into

as a team of first time hackers we ran into a couple of problems:

  1. Lack of available documents and tutorials for embedding AutoML Vision ML model into android app.
  2. most of the team did not know java
  3. most of the team did not know android development
  4. most of the team did not know how to train ML models
  5. Connecting and implementing Firebase.

Accomplishments that we're proud of

we are very proud that we were able to learn how to build something that can truly help the world in a tight timeframe. To highlight some of the thighs we are proud of.

  • we made an app as a team
  • we made an app that uses AI to recognize images
  • we made an app that can truly be used to help people save our world

What we learned

There is so much that we learned in our first hackathon

  1. java and how to use it for android development
  2. Android studio and XML
  3. how to train ML models using AutoML
  4. l how to use Firebase
  5. how to work with a team
  6. how to use GitHub as a VCS

What's next for Recyclable

We hope to add more features

  1. Realtime leaderboard updates
  2. Firebase data base to save user stats
  3. maybe a prize system for reaching certain milestone
  4. create an IOS app at some point
  5. Hopefully make a world a healthier place!

References

Data used to train ML model https://drive.google.com/drive/folders/0B3P9oO5A3RvSUW9qTG11Ul83TEE?resourcekey=0-F-D8v2tnSfByG6ll3t9JxA

Built With

Share this project:

Updates