You are here

Towards Zero Waste: Automated Waste Classification via Computer Vision (In Progress)

Recent Project Updates

View all project updates »

Project Family

Description

Funded in early 2023, this project aims to develop an automated waste characterization system that uses recent advances in computer vision to detect and classify waste more efficiently for recycling. It will also create a dashboard with live data on the nature and extent of waste generated at the U of I to motivate the campus community to follow best practices for waste disposal and recycling and to help meet zero-waste goals in the Illinois Climate Action Plan (iCAP). The campus generates nearly 5,000 tons of waste per year, and the recycling stream is manually sorted by five to seven individuals. Using cameras installed at the university’s Waste Transfer Station, the research team will develop a machine-learning model to classify waste on a moving conveyer belt into six categories — paper, plastic, food, metal, glass, and yard waste — then feed that data into the live dashboard. The project will also determine the best pathways for converting the diverse components of municipal solid waste into biofuels and bioproducts.

The Project Team

    • Nishant Garg, Assistant Professor of Civil and Environmental Engineering
    • Daphne Hulse, Zero Waste Coordinator, Facilities & Services
    • Joy Scrogum, Assistant Scientist, Sustainability, Illinois Sustainable Technology Center
    • Lav R. Varshney, Associate Professor of Electrical and Computer Engineering

Background

Purpose of the Work: Turning Waste Into Treasure

Waste management is a pervasive problem that is growing continuously with the spread of urbanization. The U.S. EPA estimates that half of municipal solid waste (MSW) ends up in landfills, contributing to significant methane emissions. There is a need for new and refined resource recovery methods that have minimal impact on climate, and the global recycling industry is demanding higher quality inputs before it will accept recycled goods. Robotic control systems with mechanical arms and machine learning for identification can sort waste more efficiently, reducing processing time and turning waste into valuable resources.

Website URL(s)

No description has been provided yet.

Project Team

  • Project Leader:

    Dr. Nishant Garg

    Team Members:

    • Joy Scrogum
    • Dr. Lav R. Varshney
    • Daphne Hulse

Dates

  • Investigated October 27, 2022
    Investigated by Dr. Nishant Garg
    Approved February 2, 2023
    Approved by iSEE

Themes

Project Location(s)

This map is interactive! Click (or touch) and drag to pan; scroll (or pinch) to zoom.

View larger location(s) map

Project details last updated on: 3/24/2023. Check Project Updates for recent activity.