Enhanced Precision in Built Environment Measurement: Integrating AprilTags Detection with Machine Learning

Shengtao Tan, Aravind Srinivasaragavan, Kepa Iturralde, Christoph Holst

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In the field of building renovation with prefabricated modules, accurately locating and identifying connectors’ positions and orientations is an essential technological challenge. For building renovation with prefabricated modules, traditional methods like total stations are not only time-consuming but also highly dependent on experienced technicians. However, previous research has proven that ApriTtag tags can be effectively used in building measurements. This paper proposes a refined AprilTag detection pipeline that integrates machine learning techniques, significantly improving detection accuracy. Moreover, this process can be easily used by non-experts making it more accessible and less time-consuming.

Original languageEnglish
Title of host publicationProceedings of the 41st International Symposium on Automation and Robotics in Construction, ISARC 2024
PublisherInternational Association for Automation and Robotics in Construction (IAARC)
Pages1295-1298
Number of pages4
ISBN (Electronic)9780645832211
DOIs
StatePublished - 2024
Event41st International Symposium on Automation and Robotics in Construction, ISARC 2024 - Lille, France
Duration: 3 Jun 20245 Jun 2024

Publication series

NameProceedings of the International Symposium on Automation and Robotics in Construction
ISSN (Electronic)2413-5844

Conference

Conference41st International Symposium on Automation and Robotics in Construction, ISARC 2024
Country/TerritoryFrance
CityLille
Period3/06/245/06/24

Keywords

  • AprilTag
  • Building Measurement
  • Machine Learning
  • Neural Network

Fingerprint

Dive into the research topics of 'Enhanced Precision in Built Environment Measurement: Integrating AprilTags Detection with Machine Learning'. Together they form a unique fingerprint.

Cite this