Skip to content
Recyclability Classifier
Java Python Jupyter Notebook CSS HTML JavaScript
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
projectWAMP-Android-RealTime
projectWAMP-Flask
projectWAMP-JupyterNotebook
projectwamp-Android-BestApp
.DS_Store
Final_WAMP_3000_tf_lite_100+50epochs.ipynb
README.md
classification_model.h5.zip
detection_labels.txt
detection_model.pb.zip
logo.png
webappdemo.png

README.md

projectWAMP

alt text

Overview

projectWAMP is our final group project in Google's Applied Machine Learning Intensive.

We intended to create a recyclability classifier that encourages good recycling practices by educating users in distinguishing between recyclable, nonrecyclable, and organic material and provides tips and guides on how to recycle and dispose different types of material.

Dataset

Click this link to download our zipped dataset file: https://drive.google.com/file/d/1a-C6BTHXAv-qmX_pfx-ocU2558JcXVlo/view?usp=sharing

Labels

The current state of our model has five labels:

  • Blue (Paper material)
  • Green (Plastic, glass beverage bottles; aluminum, steel, and tin cans)
  • Orange (Organic material)
  • Purple (Electronic)
  • Red (Nonrecyclable trash; easily contaminated material; waste that require specialized recycling methods)

Process

  1. Input Image
  2. Object Detection
  3. Crop Subimages
  4. Recyclability Classification
  5. Results

Web Application

alt text

References

Authors

Wayne Chim, Abdoul Aziz Sandotin Coulibaly, Min Henderson, Prashamsa Rimal

AMLI @ Agnes Scott Summer 2019

You can’t perform that action at this time.