Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Epidemiological and Clinical Insights from 68 Veterinarian-Reported Cases of Feline Infectious Peritonitis During the Documented FIP Epizootic in Cyprus.

Pathogens (Basel, Switzerland)·2026
Same author

Emerging applications of large language models in ecology and conservation science.

Conservation biology : the journal of the Society for Conservation Biology·2026
Same author

An ecoacoustic dataset collected on the island of Cyprus in the Mediterranean Basin biodiversity hotspot.

Scientific data·2025
Same author

Guaranteed Coverage Prediction Intervals With Gaussian Process Regression.

IEEE transactions on pattern analysis and machine intelligence·2024
Same author

CODD: A benchmark dataset for the automated sorting of construction and demolition waste.

Waste management (New York, N.Y.)·2024
Same author

Reconciling a national methane emission inventory with in-situ measurements.

The Science of the total environment·2023

Related Experiment Video

Updated: Jul 28, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

586

Real-time construction demolition waste detection using state-of-the-art deep learning methods; single-stage vs

Demetris Demetriou1, Pavlos Mavromatidis2, Ponsian M Robert2

  • 1Department of Civil & Environmental Engineering, University of Cyprus, Nicosia 1303, Cyprus.

Waste Management (New York, N.Y.)
|June 3, 2023
PubMed
Summary

The YOLOv7 deep learning model excels in accurately and rapidly identifying construction and demolition waste (CDW) components like bricks, concrete, and tiles, even when heavily stacked. This advancement is crucial for developing effective waste sorting robots.

Keywords:
Construction and Demolition WasteConvolutional neural networksDeep learningObject detectionWaste sorting

More Related Videos

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
08:27

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines

Published on: January 5, 2024

1.2K
Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

10.7K

Related Experiment Videos

Last Updated: Jul 28, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

586
Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
08:27

Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines

Published on: January 5, 2024

1.2K
Design and Analysis for Fall Detection System Simplification
08:05

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

10.7K

Area of Science:

  • Robotics and Automation
  • Computer Vision
  • Artificial Intelligence

Background:

  • Accurate object detection is vital for automated waste sorting robots.
  • Construction and Demolition Waste (CDW) presents unique challenges for real-time detection due to varied states.

Purpose of the Study:

  • To evaluate deep learning models for real-time localization and classification of CDW.
  • To identify the most effective model for waste sorting robot applications.

Main Methods:

  • Trained and tested 18 deep learning models (single-stage: SSD, YOLO; two-stage: Faster-RCNN) with various backbones (ResNet, MobileNetV2, efficientDet).
  • Utilized a novel, openly accessible CDW dataset with 6600 images of brick, concrete, and tile.
  • Evaluated models on datasets with normally and heavily stacked/adhered CDW samples.

Main Results:

  • YoloV7 achieved the highest accuracy (mAP50:95 ≈ 70%) with the fastest inference speed (<30 ms).
  • YoloV7 demonstrated robust performance on severely stacked and adhered CDW samples.
  • Faster-RCNN models showed the least accuracy fluctuations across different datasets, indicating high robustness.

Conclusions:

  • YoloV7 is the optimal choice for real-time CDW detection in waste sorting robots.
  • Further research could explore Faster-RCNN's robustness for specific challenging scenarios.