WastEd

Smart trashcan that sorts waste using machine learning

Comment

WastEd

Recognizes

  • Camera sensor detects an object has entered the WastEd sorting bin

Captures

  • Smart trashcan captures a picture of the object and sends it to the backend for processing
  • OpenCV implementation in the backend filters the image before it's processed

Identifies

  • Image is analyzed using Google Cloud Vision API
  • Suggested object type and its properties are recorded

Classifies

  • Based on the image processing results and their respective confidence levels, waste is considered for compost, recycling, or trash
  • Backend sorting algorithm designates object for appropriate category of waste

Sorts

  • Hand built hardware device comprised of sorting container and conveyor, 3-D print attachment, motor, Arduino, and standard web camera
  • Hardware contraption receives signal from backend to sort waste into selected bin
  • Sorting container moves toward appropriate bin, dumps the waste, and readjusts for more incoming trash

Updates

  • Backend writes results of trash sort to Google Cloud Firestore Database using REST API
  • Configured WastEd smart trashcans receive a comprehensive score based on input placed into compost, recycling, and trash

Compete

  • WastEd users are encouraged to produce more compost and recycling, and less trash
  • Companies and organizations will offer incentives for individuals and/or groups, depending on smart trashcan's private or public use
  • Scores update to WastEd website, built on a Flask app for python

Keep Up

  • WastEd custom website contains information about WastEd, including how it's made and where it's being used

Built With

Share this project:

Updates