Inspiration

With our world literally on fire, every contribution can make a big difference to the planet. We wanted to end the struggle of recycling by creating an app able to scan your trash and tell you where it belongs to.

What it does

The app automatically opens a bar code scanner. Then, you just have to pass the products under the scanner and all it will tell you in which bin (trash/recycling/compost) it belongs!

How we built it

The app was built using Xcode, in Swift. The scanner and computer vision was brought by Firebase made by Google. We used the free and open-source database "OpenFoodFacts" as a JSON API to retrieve the product information from the bar code identifier. We then cleaned and selected the data we wanted.

Challenges we ran into

As the first time building an app and using XCode, we had to learn a new programming language, Swift, while using new software and programming techniques. We were able to adapt ourselves and face the challenges. Implementing the bar code scanner was the most defying part. A google mentor helped us on how to implement Firebase.

Accomplishments that we're proud of

The app runs smoothly and is accessible to anyone. We have an efficient interface that makes the app easy to use. It has a lot of potentials and is a great step towards facilitating recycling.

What we learned

We learned a lot during this hackathon. Not only on a programming level (by learning how to use XCode, Swift, and building an iOS app from scratch), but also on time management and teamwork. It has been an incredible weekend, we were able to meet awesome people, mentors and discover tons of innovative projects!

What's next for re<ycle

re<ycle has a lot of potentials. First, it will be able to recognize your location and adapt the output information according to your country's regulations on recycling. Then, the app will be built on a community basis. Meaning that each user will be able to add a new product in the database and thus make the app even more reliable. Finally, re<ycle will be able to use Computer Vision and Machine Learning to recognize directly the object and its material without the need of scanning a bar code.

Built With

+ 8 more
Share this project:
×

Updates