Transformer-based deformation measurement of underground structures from a single-camera video.
Saved in:
| Title: | Transformer-based deformation measurement of underground structures from a single-camera video. |
|---|---|
| Authors: | Xu, Hao-Ruo1 (AUTHOR) haoruo.xu@connect.polyu.hk, Yin, Jia-Ning2 (AUTHOR) yinlaetitia@outlook.com, Zhang, Ning1 (AUTHOR) ning-cee.zhang@polyu.edu.hk |
| Source: | Automation in Construction. Apr2025, Vol. 172, pN.PAG-N.PAG. 1p. |
| Subjects: | Underground construction, Measurement errors, Transformer models, Feature extraction, Point cloud |
| Abstract: | Measuring the deformation of underground structures, such as excavation and tunnels, is crucial for safe construction and operation. However, conventional methods, such as inclinometer probes and total stations, are labour-intensive and time-consuming for engineers. This paper proposes a transformer-based 3D reconstruction method using single-camera video to rapidly measure structural deformations. The proposed method extracts features from video frames to reconstruct point clouds, generating centrelines and Poisson models for deformation analysis. This allows fast and precise deformation measurement at any position, outperforming traditional methods. The proposed method has achieved sub-millimetre accuracy in small-scale inclinometer casings, and a 5 cm accuracy level in a large-scale tunnel, confirming its capability for detecting subtle deformations. Discrepancies in accuracy for larger structures were attributed to limitations in camera resolution, suggesting that employing 100-megapixel cameras could guarantee millimetre-level accuracy. The method's simplicity and adaptability demonstrate its potential as a practical supplement to existing deformation measurement methods. • A transformer-based method for automatic deformation measurement of underground structures is developed. • The U-Net and Transformer are combined for reliable and repeatable feature extraction from a single-camera video. • The small-diameter inclinometer casing and large-diameter tunnel are applied for validation. • The proposed method achieves satisfying measurement errors of less than 1 %. [ABSTRACT FROM AUTHOR] |
| Copyright of Automation in Construction is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
| FullText | Text: Availability: 0 |
|---|---|
| Header | DbId: egs DbLabel: Engineering Source An: 183241688 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Transformer-based deformation measurement of underground structures from a single-camera video. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Xu%2C+Hao-Ruo%22">Xu, Hao-Ruo</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> haoruo.xu@connect.polyu.hk</i><br /><searchLink fieldCode="AR" term="%22Yin%2C+Jia-Ning%22">Yin, Jia-Ning</searchLink><relatesTo>2</relatesTo> (AUTHOR)<i> yinlaetitia@outlook.com</i><br /><searchLink fieldCode="AR" term="%22Zhang%2C+Ning%22">Zhang, Ning</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> ning-cee.zhang@polyu.edu.hk</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Automation+in+Construction%22">Automation in Construction</searchLink>. Apr2025, Vol. 172, pN.PAG-N.PAG. 1p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Underground+construction%22">Underground construction</searchLink><br /><searchLink fieldCode="DE" term="%22Measurement+errors%22">Measurement errors</searchLink><br /><searchLink fieldCode="DE" term="%22Transformer+models%22">Transformer models</searchLink><br /><searchLink fieldCode="DE" term="%22Feature+extraction%22">Feature extraction</searchLink><br /><searchLink fieldCode="DE" term="%22Point+cloud%22">Point cloud</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Measuring the deformation of underground structures, such as excavation and tunnels, is crucial for safe construction and operation. However, conventional methods, such as inclinometer probes and total stations, are labour-intensive and time-consuming for engineers. This paper proposes a transformer-based 3D reconstruction method using single-camera video to rapidly measure structural deformations. The proposed method extracts features from video frames to reconstruct point clouds, generating centrelines and Poisson models for deformation analysis. This allows fast and precise deformation measurement at any position, outperforming traditional methods. The proposed method has achieved sub-millimetre accuracy in small-scale inclinometer casings, and a 5 cm accuracy level in a large-scale tunnel, confirming its capability for detecting subtle deformations. Discrepancies in accuracy for larger structures were attributed to limitations in camera resolution, suggesting that employing 100-megapixel cameras could guarantee millimetre-level accuracy. The method's simplicity and adaptability demonstrate its potential as a practical supplement to existing deformation measurement methods. • A transformer-based method for automatic deformation measurement of underground structures is developed. • The U-Net and Transformer are combined for reliable and repeatable feature extraction from a single-camera video. • The small-diameter inclinometer casing and large-diameter tunnel are applied for validation. • The proposed method achieves satisfying measurement errors of less than 1 %. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of Automation in Construction is the property of Elsevier B.V. and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=183241688 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1016/j.autcon.2025.106070 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 1 StartPage: N.PAG Subjects: – SubjectFull: Underground construction Type: general – SubjectFull: Measurement errors Type: general – SubjectFull: Transformer models Type: general – SubjectFull: Feature extraction Type: general – SubjectFull: Point cloud Type: general Titles: – TitleFull: Transformer-based deformation measurement of underground structures from a single-camera video. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Xu, Hao-Ruo – PersonEntity: Name: NameFull: Yin, Jia-Ning – PersonEntity: Name: NameFull: Zhang, Ning IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 04 Text: Apr2025 Type: published Y: 2025 Identifiers: – Type: issn-print Value: 09265805 Numbering: – Type: volume Value: 172 Titles: – TitleFull: Automation in Construction Type: main |
| ResultId | 1 |
