Application Framework and Optimal Features for UAV-Based Earthquake-Induced Structural Displacement Monitoring

Unmanned aerial vehicle (UAV) vision-based sensing has become an emerging technology for structural health monitoring (SHM) and post-disaster damage assessment of civil infrastructure. This article proposes a framework for monitoring structural displacement under earthquakes by reprojecting image po...

Full description

Saved in:
Bibliographic Details
Main Authors: Ruipu Ji, Shokrullah Sorosh, Eric Lo, Tanner J. Norton, John W. Driscoll, Falko Kuester, Andre R. Barbosa, Barbara G. Simpson, Tara C. Hutchinson
Format: Article
Language:English
Published: MDPI AG 2025-01-01
Series:Algorithms
Subjects:
Online Access:https://www.mdpi.com/1999-4893/18/2/66
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849722501873008640
author Ruipu Ji
Shokrullah Sorosh
Eric Lo
Tanner J. Norton
John W. Driscoll
Falko Kuester
Andre R. Barbosa
Barbara G. Simpson
Tara C. Hutchinson
author_facet Ruipu Ji
Shokrullah Sorosh
Eric Lo
Tanner J. Norton
John W. Driscoll
Falko Kuester
Andre R. Barbosa
Barbara G. Simpson
Tara C. Hutchinson
author_sort Ruipu Ji
collection DOAJ
description Unmanned aerial vehicle (UAV) vision-based sensing has become an emerging technology for structural health monitoring (SHM) and post-disaster damage assessment of civil infrastructure. This article proposes a framework for monitoring structural displacement under earthquakes by reprojecting image points obtained courtesy of UAV-captured videos to the 3-D world space based on the world-to-image point correspondences. To identify optimal features in the UAV imagery, geo-reference targets with various patterns were installed on a test building specimen, which was then subjected to earthquake shaking. A feature point tracking-based algorithm for square checkerboard patterns and a Hough Transform-based algorithm for concentric circular patterns are developed to ensure reliable detection and tracking of image features. Photogrammetry techniques are applied to reconstruct the 3-D world points and extract structural displacements. The proposed methodology is validated by monitoring the displacements of a full-scale 6-story mass timber building during a series of shake table tests. Reasonable accuracy is achieved in that the overall root-mean-square errors of the tracking results are at the millimeter level compared to ground truth measurements from analog sensors. Insights on optimal features for monitoring structural dynamic response are discussed based on statistical analysis of the error characteristics for the various reference target patterns used to track the structural displacements.
format Article
id doaj-art-0c1a547610da4e9ba41f4fc3308ade7f
institution DOAJ
issn 1999-4893
language English
publishDate 2025-01-01
publisher MDPI AG
record_format Article
series Algorithms
spelling doaj-art-0c1a547610da4e9ba41f4fc3308ade7f2025-08-20T03:11:19ZengMDPI AGAlgorithms1999-48932025-01-011826610.3390/a18020066Application Framework and Optimal Features for UAV-Based Earthquake-Induced Structural Displacement MonitoringRuipu Ji0Shokrullah Sorosh1Eric Lo2Tanner J. Norton3John W. Driscoll4Falko Kuester5Andre R. Barbosa6Barbara G. Simpson7Tara C. Hutchinson8Department of Structural Engineering, University of California San Diego, La Jolla, CA 92093, USADepartment of Structural Engineering, University of California San Diego, La Jolla, CA 92093, USAQualcomm Institute, University of California San Diego, La Jolla, CA 92093, USAQualcomm Institute, University of California San Diego, La Jolla, CA 92093, USAQualcomm Institute, University of California San Diego, La Jolla, CA 92093, USADepartment of Structural Engineering, University of California San Diego, La Jolla, CA 92093, USASchool of Civil and Construction Engineering, Oregon State University, Corvallis, OR 97331, USADepartment of Civil and Environmental Engineering, Stanford University, Stanford, CA 94305, USADepartment of Structural Engineering, University of California San Diego, La Jolla, CA 92093, USAUnmanned aerial vehicle (UAV) vision-based sensing has become an emerging technology for structural health monitoring (SHM) and post-disaster damage assessment of civil infrastructure. This article proposes a framework for monitoring structural displacement under earthquakes by reprojecting image points obtained courtesy of UAV-captured videos to the 3-D world space based on the world-to-image point correspondences. To identify optimal features in the UAV imagery, geo-reference targets with various patterns were installed on a test building specimen, which was then subjected to earthquake shaking. A feature point tracking-based algorithm for square checkerboard patterns and a Hough Transform-based algorithm for concentric circular patterns are developed to ensure reliable detection and tracking of image features. Photogrammetry techniques are applied to reconstruct the 3-D world points and extract structural displacements. The proposed methodology is validated by monitoring the displacements of a full-scale 6-story mass timber building during a series of shake table tests. Reasonable accuracy is achieved in that the overall root-mean-square errors of the tracking results are at the millimeter level compared to ground truth measurements from analog sensors. Insights on optimal features for monitoring structural dynamic response are discussed based on statistical analysis of the error characteristics for the various reference target patterns used to track the structural displacements.https://www.mdpi.com/1999-4893/18/2/66unmanned aerial vehicle (UAV)vision-based sensingearthquaketarget trackingworld point reconstructionstructural displacement monitoring
spellingShingle Ruipu Ji
Shokrullah Sorosh
Eric Lo
Tanner J. Norton
John W. Driscoll
Falko Kuester
Andre R. Barbosa
Barbara G. Simpson
Tara C. Hutchinson
Application Framework and Optimal Features for UAV-Based Earthquake-Induced Structural Displacement Monitoring
Algorithms
unmanned aerial vehicle (UAV)
vision-based sensing
earthquake
target tracking
world point reconstruction
structural displacement monitoring
title Application Framework and Optimal Features for UAV-Based Earthquake-Induced Structural Displacement Monitoring
title_full Application Framework and Optimal Features for UAV-Based Earthquake-Induced Structural Displacement Monitoring
title_fullStr Application Framework and Optimal Features for UAV-Based Earthquake-Induced Structural Displacement Monitoring
title_full_unstemmed Application Framework and Optimal Features for UAV-Based Earthquake-Induced Structural Displacement Monitoring
title_short Application Framework and Optimal Features for UAV-Based Earthquake-Induced Structural Displacement Monitoring
title_sort application framework and optimal features for uav based earthquake induced structural displacement monitoring
topic unmanned aerial vehicle (UAV)
vision-based sensing
earthquake
target tracking
world point reconstruction
structural displacement monitoring
url https://www.mdpi.com/1999-4893/18/2/66
work_keys_str_mv AT ruipuji applicationframeworkandoptimalfeaturesforuavbasedearthquakeinducedstructuraldisplacementmonitoring
AT shokrullahsorosh applicationframeworkandoptimalfeaturesforuavbasedearthquakeinducedstructuraldisplacementmonitoring
AT ericlo applicationframeworkandoptimalfeaturesforuavbasedearthquakeinducedstructuraldisplacementmonitoring
AT tannerjnorton applicationframeworkandoptimalfeaturesforuavbasedearthquakeinducedstructuraldisplacementmonitoring
AT johnwdriscoll applicationframeworkandoptimalfeaturesforuavbasedearthquakeinducedstructuraldisplacementmonitoring
AT falkokuester applicationframeworkandoptimalfeaturesforuavbasedearthquakeinducedstructuraldisplacementmonitoring
AT andrerbarbosa applicationframeworkandoptimalfeaturesforuavbasedearthquakeinducedstructuraldisplacementmonitoring
AT barbaragsimpson applicationframeworkandoptimalfeaturesforuavbasedearthquakeinducedstructuraldisplacementmonitoring
AT tarachutchinson applicationframeworkandoptimalfeaturesforuavbasedearthquakeinducedstructuraldisplacementmonitoring