Waste Finder

• 

  Example of our Website

Inspiration

Time. Time inspired us to make sure that our future generations aren't taken over with plastic waste. We want the future of our world to be devoid of plastic waste.

What it does

Our application marks a google map with different concentration levels of plastic waste that is reported to us by people like you. When a person encounters trash, they would simply take a photo of it and we would detect the amount of plastic waste (specifically plastic bottles) in the photo. If there are a lot, then we would classify the place with a high concentration of plastic waste and if not then a low concentration. This brings awareness to the plastic waste issue and people will be able to take action and go to those places to pick up trash.

How we built it

We used Python flask for URL routing and backend to store data, and HTML/CSS/JS for frontend and also OpenCV with YOLO trained on a COCO dataset for plastic bottle image detection, also we use the Google Maps API

Challenges we ran into

We didn't know how to work with OpenCV until after the CV workshop yesterday, and it was hard trying to integrate the project together in such a short period of time. We also weren't sure how to link up a heat map with google maps based on wherever the person posts the picture.

Accomplishments that we're proud of

We're proud how we were able to detect plastic bottles with OpenCV and how we made an entire website based on that premise

What we learned

We learned about ML, CV, and web development. Also we learned about the lifestyle of working in a team to solve a problem that we had no idea how to solve at first.

What's next for Waste Finder

Globalizing to other cities based on location and estimating price of plastic bottles and creating events for people to go to and implementing a search function for easy access to plastic bottle sites

Built With

• coco
• css3
• flask
• google-maps
• html5
• javascript
• json
• mediadevices
• opencv
• python
• yolo