Related Experiment Video
Updated: May 2, 2026
08:25
Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
Published on: May 7, 2019
9.1K
A Bottom-Up Approach Integrating Computer Vision with Material Flow Analysis to Estimate the Recycling Potential of
Peijin Jiang1, Hanwen Xu2, Qingshi Tu1
1Sustainable Bioeconomy Research Group, Department of Wood Science, The University of British Columbia, Vancouver, British Columbia V6T 1Z4, Canada.
Environmental Science & Technology
|August 22, 2025
Summary
Managing end-of-life solar photovoltaic (PV) panels is a growing challenge. SolarScope, an open-source model, uses computer vision to identify PV installations and assess recycling potential, aiding circular economy goals.
Area of Science:
- Environmental Science
- Materials Science
- Computer Science
Background:
- Rapid solar photovoltaic (PV) deployment necessitates effective end-of-life management strategies.
- Existing recycling potential studies are limited by insufficient data on distributed PV installations.
- Accurate mapping of PV systems is crucial for urban mining and circular economy initiatives.
Purpose of the Study:
- To develop an automated method for identifying and quantifying distributed solar PV installations.
- To evaluate the urban mining potential of end-of-life PV panels using integrated modeling.
- To provide a scalable solution for material flow analysis in the context of renewable energy systems.
Main Methods:
- Integration of computer vision (CV) techniques with dynamic material flow analysis (dMFA).
- Utilized satellite imagery and Vision Transformer (ViT) models for PV panel classification and segmentation.
- Developed an open-source model, SolarScope, for automated PV identification and potential assessment.
Main Results:
- SolarScope achieved high accuracy in PV installation classification (AUROC=0.93) and segmentation (Dice=0.90).
- Demonstrated model transferability and effectiveness in fine-scale material recovery assessments via a case study in Kamakura, Japan.
- Successfully combined CV and dMFA to bottom-up estimate regional material stock and recycling potential.
Conclusions:
- The SolarScope model offers a scalable solution to overcome data limitations in conventional material flow analysis for distributed PV systems.
- This methodological framework supports enhanced urban mining and promotes the advancement of the circular economy in the renewable energy sector.
- Accurate, automated identification of PV installations is key to sustainable end-of-life management and resource recovery.
