Meerwald, Taylor (2023) Smart Condition Assessment of Tunnel Structures. [USQ Project]
![]() |
Text (Project – redacted)
Meerwald_T_Ngyuen_Redacted.pdf Download (4MB) |
Abstract
Tunnel structures are a major part of our transport infrastructure. These tunnels have a significant financial impact with approximately $9.5 billion spent on their construction and $105 million spent on maintenance in the 2021-2022 financial year(Moore 2019; Transurban 2022). Part of the maintenance processes is the annual inspection of the tunnel structure(Louis 2018). Currently this is done by an experienced engineer who physically examines the tunnel. Recent research has shown that artificial intelligence (AI) technologies can be used to assist with the detection of defects in structures. This paper aims to compare the effect of lighting and resolution quality on the training of a machine learning model for tunnel condition assessment applications.
The project found that there are a number of complicated and interacting variables to navigate in a tunnel environment. It showed that a higher resolution and lighting quality was beneficial in the annotation of images for training of a defect detection model. However, further data would need ot be collected and annotated, ensuring that all classes and quality types have the same number of instances so that the only variable between models is the quality of the image. Whilst this project has not directly contributed to the ultimate industry goal for an integrated infield drone controller and live in-field inspection device, it has helped to gain an understanding of the effects of the challenging variables that are encountered in a tunnel environment.
![]() |
Statistics for this ePrint Item |
Item Type: | USQ Project |
---|---|
Item Status: | Live Archive |
Faculty/School / Institute/Centre: | Current – Faculty of Health, Engineering and Sciences - School of Engineering (1 Jan 2022 -) |
Supervisors: | Ngyuen, Andy |
Qualification: | Bachelor of Engineering (Honours) (Civil) |
Date Deposited: | 30 Sep 2025 03:04 |
Last Modified: | 30 Sep 2025 03:04 |
Uncontrolled Keywords: | tunnels; AI technology; defects |
URI: | https://sear.unisq.edu.au/id/eprint/52972 |
Actions (login required)
![]() |
Archive Repository Staff Only |