Recycle Cam

Inspiration

The current climate crisis has been haunting us all, and while many try, little play a proactive role in helping the current situation. We hoped to build an app to help people, help the world.

What it does

Recycle Cam is an app with a built-in camera, which takes a picture of a waste item, they need characterized. The built-in AI model, built with Microsoft's Azur's Custom Vision, classifies the waste item as garbage, recycling, compost, or e-waste. It also includes a map feature which informs the user of nearby recycling stations/plants.

How we built it

Our team utilized Microsoft's Azur API to construct the machine learning model. The frontend website was written in React.js, which allowed for a better user interface, and the map feature used Google's Map API to locate local recycling plants.

Challenges we ran into

Some of the main challenges we ran into, was when we tried to run the machine learning model, on the frontend website. The program was not able to process the output in the correct file, instead, outputting base64 images. We had to process the base64 images, and convert them into png files.

Accomplishments that we're proud of

We were able to pull together as a team during the end when a majority of our problems seemed to dawn upon us.

What we learned

We were able to learn new technologies, such as how to train AI models with Azure's Custom Vision, and how to utilize these models in live websites.

What's next for Recycle Cam:

We hope to add in more features, like the location feature, and utilize larger datasets to create more accurate and complex AI models.

Built With

• azure
• computervision
• customvision
• javascript
• react