Development of advanced progress recognition algorithms for construction monitoring

Lai Yingdong, Lin Zhijun, Ye Zhijie and Zhang Jun

PLOS ONE, 2025, vol. 20, issue 10, 1-16

Abstract: Introduction: The traditional methods of Construction Progress Monitoring (CPM) involve manual inspection and reporting, which are slow, error-prone, and labor-intensive. Purpose: This study aims to introduce a novel, automated approach for CPM using YOLOv8, a state-of-the-art object detection algorithm, to enhance efficiency and accuracy in monitoring construction projects. Methodology: YOLOv8 is employed for its real-time processing capabilities and high precision, making it suitable for identifying and tracking construction elements in images and videos captured on-site. This study creates a comprehensive dataset of construction images and videos to assess and validate the proposed method with meticulous labeling of relevant objects. Results: A custom-labeled dataset of 768 images of window installation stages was developed and used to train the model. The proposed YOLOv8 model achieved a mean Average Precision (mAP@50) of 0.953, mAP@50–95 of 0.678, precision of 0.91, and recall of 0.88. This integration of computer vision into CPM offers substantial benefits, including reliable, efficient, and cost-effective progress monitoring. Innovation: This approach presents an innovative computer vision application in construction progress monitoring. It facilitates timely decision-making throughout the project lifecycle and offers a practical alternative to manual CPM methods. Using YOLOv8 for automated CPM is a novel contribution to construction project management, potentially impacting the successful completion.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0333262 (text/html)
https://journals.plos.org/plosone/article/file?id= ... 33262&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pone00:0333262

DOI: 10.1371/journal.pone.0333262

Access Statistics for this article

More articles in PLOS ONE from Public Library of Science
Bibliographic data for series maintained by plosone ().