Buy article PDF
The purchased file will be sent to you
via email after the payment is completed.
US$ 35
|
Smart Structures and Systems Volume 33, Number 5, May 2024 , pages 349-358 DOI: https://doi.org/10.12989/sss.2024.33.5.349 |
|
|
|
Loop closure-based high-resolution façade digital modeling technique of large-scale dams using UAV |
||
Myung Soo Kang, Keunyoung Jang, Yong-Rae Yu and Yun-Kyu An
|
||
| Abstract | ||
| Structural digital models can be effectively established by spatially obtaining digital images using an unmanned aerial vehicle (UAV). One of the main purposes of the structural digital modeling is computer vision-based exterior damage detection of a target structure. To investigate micro-scale damage from the digital model, high-resolution digital images obtained with a close-up vision survey is typically required. However, serial image synthesis such as image stitching may cumulate stitching errors as the number of scanned images increases. Therefore, in this paper, a novel loop closure-based digital image stitching technique is proposed and experimentally validated using the close-up surveyed digital images acquired from an in-situ dam structure located in South Korea. The test results reveal that the proposed technique outperforms a non-loop closure-based image stitching technique, which can cause serious distortions, such as ghosting and vanishing phenomena. | ||
| Key Words | ||
| close-up UAV surveying; computer vision; dam; digital image stitching; loop closure; structural façade digital model | ||
| Address | ||
| (1) Myung Soo Kang, Keunyoung Jang, Yun-Kyu An: Department of Architectural Engineering, Sejong University, 209 Neungdong-ro, Gwangjin-gu, Seoul 05006, Republic of Korea; (2) Yong-Rae Yu: Department of Civil and Environmental Engineering, Sejong University, 209 Neungdong-ro, Gwangjin-gu, Seoul 05006, Republic of Korea. | ||