UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Cascade refinement extraction network with active boundary loss for segmentation of concrete cracks from high-resolution images

Deng, L; Yuan, H; Long, L; Chun, PJ; Chen, W; Chu, H; (2024) Cascade refinement extraction network with active boundary loss for segmentation of concrete cracks from high-resolution images. Automation in Construction , 162 , Article 105410. 10.1016/j.autcon.2024.105410. Green open access

[thumbnail of 1-s2.0-S0926580524001468-main.pdf]
Preview
PDF
1-s2.0-S0926580524001468-main.pdf - Published Version

Download (20MB) | Preview

Abstract

Accurate extraction of cracks is important yet challenging in bridge inspection, particularly that of tiny cracks captured from high-resolution (HR) images. This paper presents a crack-boundary refinement framework (CBRF) for meticulous segmentation of HR crack images. First, a triple-scale feature extraction module is designed to enhance the representation of miniscule-crack pixels. Then, a cascade operation involving global and local steps is adopted to conduct the refinement. In addition, an active boundary loss is introduced into the training process to solve the semantic inconsistency of crack boundary areas. The first HR crack image dataset is established to thoroughly evaluate the CBRF. Finally, an unmanned aerial vehicle (UAV)-based case study is conducted on the Yinpenling Bridge, which further confirms the practicality of the CBRF in improving the safety and efficiency of UAV-based bridge detection while ensuring the accuracy.

Type: Article
Title: Cascade refinement extraction network with active boundary loss for segmentation of concrete cracks from high-resolution images
Open access status: An open access version is available from UCL Discovery
DOI: 10.1016/j.autcon.2024.105410
Publisher version: http://dx.doi.org/10.1016/j.autcon.2024.105410
Language: English
Additional information: This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Keywords: Deep learning, Crack segmentation, High resolution dataset, Multi-scale cascade operation, Boundary refinement, UAV bridge inspection
UCL classification: UCL
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of the Built Environment
URI: https://discovery.ucl.ac.uk/id/eprint/10191738
Downloads since deposit
9Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item