• Skip to primary navigation
  • Skip to main content
  • About
  • Contact
  • Listserv
  • Facebook
  • Twitter
  • Login/Register
workzonesafety.org

National Work Zone Safety Information Clearinghouse

Library of Resources to Improve Roadway Work Zone Safety for All Roadway Users

  • Crash Information
    • Work Zone Fatal Crashes and Fatalities
    • National Estimates of Total and Injury Work Zone Crashes
    • Work Zone Fatal Crash Facts
    • Other Work Zone Crash Related Resources
  • Flagger Information
    • Flagger Training and Certification Requirements by State
    • Flagging Resources Developed Under the FHWA Work Zone Safety Grant Program
    • Classes
    • Videos
    • Job Resources
  • Training
    • FHWA Work Zone Safety Grant Products
    • Roadway Safety Training Program
    • Online Learning Management System
    • Training Courses and Programs
    • Training Videos
    • ARTBA
    • ATSSA
    • NHI
    • NSC
    • LTAP/TTAP
    • WSU
  • Events and Conferences
    • National Work Zone Awareness Week 2021
    • Work Zone Safety Conferences
    • Other Work Zone Related Conferences
    • 2021 Transportation Construction Safety Events Calendar
  • Data Resources
    • Searchable Databases
    • Public Awareness
    • Laws, Regulations, Standards, and Policies
    • Manual on Uniform Traffic Control Devices (MUTCD)
  • Hot Topics
    • Improving Large Truck Safety in Work Zones
    • Smarter Work Zones
    • Transportation Management Plans
    • Accommodating Pedestrians
    • Rear End Crashes
    • Minimizing Construction Interferences with Traffic
    • Managing Speeds
    • Mobile/Short Duration Work Zones
    • Reducing Worker Run-overs and Back-overs
    • Working at Night
    • Reducing Worker Fatigue and Distraction Risks
You are here: Home / Publications / Human-Object Interaction Recognition for Automatic Construction Site Safety Inspection

Human-Object Interaction Recognition for Automatic Construction Site Safety Inspection

Author/Presenter: Tang, Shuai; Roberts, Dominic; Golparvar-Fard, Mani
Abstract:

Today, computer vision object detection methods are used for safety inspections from site videos and images. These methods detect bounding boxes and use hand-made rules to enable personal protective equipment compliance checks. This paper presents a new method to improve the breadth and depth of vision-based safety compliance checking by explicitly classifying worker-tool interactions. A detection model is trained on a newly constructed image dataset for construction sites, achieving 52.9% average mean precision for 10 object categories and 89.4% average precision for detecting workers. Using this detector and new dataset, the proposed human-object interaction recognition model achieved 79.78% precision and 77.64% recall for hard hat checking; 79.11% precision and 75.29% recall for safety coloring checking. The new model also verifies hand protection for workers when tools are being used with 66.2% precision and 64.86% recall. The proposed model is superior in these checking tasks when compared with post-processing detected objects with hand-made rules, or applying detected objects only.

Source: Automation in Construction
Volume: 120
Publication Date: December 2020
Source URL: Link to URL
Publication Types: Books, Reports, Papers, and Research Articles
Topics: Computer Vision; Construction Safety; Detection and Identification Technologies; Inspection; Personal Protective Equipment; Worker Safety

Copyright © 2021 American Road & Transportation Builders Association (ARTBA). The National Work Zone Safety Information Clearinghouse is a project of the ARTBA Transportation Development Foundation. It is operated in cooperation with the U.S. Federal Highway Administration and Texas A&M Transportation Institute. | Copyright Statement · Legal Notices/Policies · Disclaimer
American Road and Transportation Builders Association Transportation Development Foundation, American Road and Transportation Builders Association U.S. Department of Transportation Federal Highway Administration Texas A&M Transportation Institute