High-precision dimensional measurement of a curtain wall cross-section using image super-resolution

Jun Su Park; Hyo Seon Park

doi:10.21741/9781644902455-28

High-precision dimensional measurement of a curtain wall cross-section using image super-resolution

Jun Su Park, Hyo Seon Park

Department of Architecture and Architectural Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Dimensional quality is critical to the successful installation of curtain walls, and the required dimensional tolerances are typically less than a millimeter. However, high-precision dimensional measurement of a curtain wall cross-section is practically difficult and time-consuming because the cross-sectional shapes are various and complicated, and dimensional measurement is usually performed manually in the actual field. To improve these problems, various vision-based methos are being attempted, but there have been limitations in terms of precision due to row image resolution. Therefore, this study confirmed whether image super-resolution can contribute to overcoming these limitations. To this end, an experiment on a curtain wall profile cross-section was conducted, and super-resolution generative adversarial network (SRGAN) was applied as an image super-resolution method. As a result, it was confirmed that high-precision dimensional measurement is possible from an image with enhanced resolution using SRGAN.

Original language	English
Title of host publication	Structural Health Monitoring- The 9th Asia-Pacific Workshop on Structural Health Monitoring, 9APWSHM 2022
Editors	Nik Rajic, Wing Kong Chiu, Martin Veidt, Akira Mita, N. Takeda
Publisher	Association of American Publishers
Pages	223-227
Number of pages	5
ISBN (Print)	9781644902448
DOIs	https://doi.org/10.21741/9781644902455-28
Publication status	Published - 2023
Event	9th Asia-Pacific Workshop on Structural Health Monitoring, 9APWSHM 2022 - Cairns, Australia Duration: 2022 Dec 7 → 2022 Dec 9

Publication series

Name	Materials Research Proceedings
Volume	27
ISSN (Print)	2474-3941
ISSN (Electronic)	2474-395X

Conference

Conference	9th Asia-Pacific Workshop on Structural Health Monitoring, 9APWSHM 2022
Country/Territory	Australia
City	Cairns
Period	22/12/7 → 22/12/9

Bibliographical note

Publisher Copyright:
© 2023, Association of American Publishers. All rights reserved.

All Science Journal Classification (ASJC) codes

General Materials Science

Access to Document

10.21741/9781644902455-28

Cite this

Park, J. S., & Park, H. S. (2023). High-precision dimensional measurement of a curtain wall cross-section using image super-resolution. In N. Rajic, W. K. Chiu, M. Veidt, A. Mita, & N. Takeda (Eds.), Structural Health Monitoring- The 9th Asia-Pacific Workshop on Structural Health Monitoring, 9APWSHM 2022 (pp. 223-227). (Materials Research Proceedings; Vol. 27). Association of American Publishers. https://doi.org/10.21741/9781644902455-28

Park, Jun Su ; Park, Hyo Seon. / High-precision dimensional measurement of a curtain wall cross-section using image super-resolution. Structural Health Monitoring- The 9th Asia-Pacific Workshop on Structural Health Monitoring, 9APWSHM 2022. editor / Nik Rajic ; Wing Kong Chiu ; Martin Veidt ; Akira Mita ; N. Takeda. Association of American Publishers, 2023. pp. 223-227 (Materials Research Proceedings).

@inproceedings{3d27e5db513049f9a4605bf8815314b7,

title = "High-precision dimensional measurement of a curtain wall cross-section using image super-resolution",

abstract = "Dimensional quality is critical to the successful installation of curtain walls, and the required dimensional tolerances are typically less than a millimeter. However, high-precision dimensional measurement of a curtain wall cross-section is practically difficult and time-consuming because the cross-sectional shapes are various and complicated, and dimensional measurement is usually performed manually in the actual field. To improve these problems, various vision-based methos are being attempted, but there have been limitations in terms of precision due to row image resolution. Therefore, this study confirmed whether image super-resolution can contribute to overcoming these limitations. To this end, an experiment on a curtain wall profile cross-section was conducted, and super-resolution generative adversarial network (SRGAN) was applied as an image super-resolution method. As a result, it was confirmed that high-precision dimensional measurement is possible from an image with enhanced resolution using SRGAN.",

keywords = "Curtain Wall, High-Precision Dimensional Measurement, Super-Resolution",

author = "Park, {Jun Su} and Park, {Hyo Seon}",

note = "Publisher Copyright: {\textcopyright} 2023, Association of American Publishers. All rights reserved.; 9th Asia-Pacific Workshop on Structural Health Monitoring, 9APWSHM 2022 ; Conference date: 07-12-2022 Through 09-12-2022",

year = "2023",

doi = "10.21741/9781644902455-28",

language = "English",

isbn = "9781644902448",

series = "Materials Research Proceedings",

publisher = "Association of American Publishers",

pages = "223--227",

editor = "Nik Rajic and Chiu, {Wing Kong} and Martin Veidt and Akira Mita and N. Takeda",

booktitle = "Structural Health Monitoring- The 9th Asia-Pacific Workshop on Structural Health Monitoring, 9APWSHM 2022",

address = "United States",

}

Park, JS & Park, HS 2023, High-precision dimensional measurement of a curtain wall cross-section using image super-resolution. in N Rajic, WK Chiu, M Veidt, A Mita & N Takeda (eds), Structural Health Monitoring- The 9th Asia-Pacific Workshop on Structural Health Monitoring, 9APWSHM 2022. Materials Research Proceedings, vol. 27, Association of American Publishers, pp. 223-227, 9th Asia-Pacific Workshop on Structural Health Monitoring, 9APWSHM 2022, Cairns, Australia, 22/12/7. https://doi.org/10.21741/9781644902455-28

High-precision dimensional measurement of a curtain wall cross-section using image super-resolution. / Park, Jun Su; Park, Hyo Seon.
Structural Health Monitoring- The 9th Asia-Pacific Workshop on Structural Health Monitoring, 9APWSHM 2022. ed. / Nik Rajic; Wing Kong Chiu; Martin Veidt; Akira Mita; N. Takeda. Association of American Publishers, 2023. p. 223-227 (Materials Research Proceedings; Vol. 27).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - High-precision dimensional measurement of a curtain wall cross-section using image super-resolution

AU - Park, Jun Su

AU - Park, Hyo Seon

PY - 2023

Y1 - 2023

N2 - Dimensional quality is critical to the successful installation of curtain walls, and the required dimensional tolerances are typically less than a millimeter. However, high-precision dimensional measurement of a curtain wall cross-section is practically difficult and time-consuming because the cross-sectional shapes are various and complicated, and dimensional measurement is usually performed manually in the actual field. To improve these problems, various vision-based methos are being attempted, but there have been limitations in terms of precision due to row image resolution. Therefore, this study confirmed whether image super-resolution can contribute to overcoming these limitations. To this end, an experiment on a curtain wall profile cross-section was conducted, and super-resolution generative adversarial network (SRGAN) was applied as an image super-resolution method. As a result, it was confirmed that high-precision dimensional measurement is possible from an image with enhanced resolution using SRGAN.

AB - Dimensional quality is critical to the successful installation of curtain walls, and the required dimensional tolerances are typically less than a millimeter. However, high-precision dimensional measurement of a curtain wall cross-section is practically difficult and time-consuming because the cross-sectional shapes are various and complicated, and dimensional measurement is usually performed manually in the actual field. To improve these problems, various vision-based methos are being attempted, but there have been limitations in terms of precision due to row image resolution. Therefore, this study confirmed whether image super-resolution can contribute to overcoming these limitations. To this end, an experiment on a curtain wall profile cross-section was conducted, and super-resolution generative adversarial network (SRGAN) was applied as an image super-resolution method. As a result, it was confirmed that high-precision dimensional measurement is possible from an image with enhanced resolution using SRGAN.

KW - Curtain Wall

KW - High-Precision Dimensional Measurement

KW - Super-Resolution

UR - http://www.scopus.com/inward/record.url?scp=85159561718&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85159561718&partnerID=8YFLogxK

U2 - 10.21741/9781644902455-28

DO - 10.21741/9781644902455-28

M3 - Conference contribution

AN - SCOPUS:85159561718

SN - 9781644902448

T3 - Materials Research Proceedings

SP - 223

EP - 227

BT - Structural Health Monitoring- The 9th Asia-Pacific Workshop on Structural Health Monitoring, 9APWSHM 2022

A2 - Rajic, Nik

A2 - Chiu, Wing Kong

A2 - Veidt, Martin

A2 - Mita, Akira

A2 - Takeda, N.

PB - Association of American Publishers

T2 - 9th Asia-Pacific Workshop on Structural Health Monitoring, 9APWSHM 2022

Y2 - 7 December 2022 through 9 December 2022

ER -

Park JS, Park HS. High-precision dimensional measurement of a curtain wall cross-section using image super-resolution. In Rajic N, Chiu WK, Veidt M, Mita A, Takeda N, editors, Structural Health Monitoring- The 9th Asia-Pacific Workshop on Structural Health Monitoring, 9APWSHM 2022. Association of American Publishers. 2023. p. 223-227. (Materials Research Proceedings). doi: 10.21741/9781644902455-28

High-precision dimensional measurement of a curtain wall cross-section using image super-resolution

Abstract

Publication series

Conference

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this