Applying Computer Vision to Reduce Contamination in the Recycling Stream
summary
Compology deploys in-dumpster cameras that upload images to a cloud pipeline using a convolutional neural network plus human review to count and time-stamp contaminants and measure fullness. Contamination is scored with a simple C-score: contaminants per volume (counts ÷ disposed volume) so haulers get alerts, targeted outreach, or fines. Deployed on about 2,000 Waste Management containers, the system reports a 57% contamination reduction over six months and a rise in completely clean pickups.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 0:27MOB 4000
- 1:20Visibility
- 3:20What is recycling
- 3:55The good news
- 5:39Our technology
- 7:04Our metric
- 8:40CScore
- 9:08CScore Examples
- 9:40CScore Process
- 10:50Real World Example
- 11:41Results
- 12:53Outro
related talks
31mFinding trash in waste
50mSchrödinger’s Warehouse: Reducing Incidents with Computer Vision
39mImpacting Global Policy by Understanding Litter Data
27mComputer vision for automated quality inspection
36mAI-Powered Real-Time Quality Control
39mPredicting failure & automating defect detection on utility assets
41mPractical Aspects of Applying Modern Computer Vision Technologies
23mComputer Vision Toolbox - Connected Computer Vision & AI Quality Inspection at Scale!