A novel directional lighting algorithm for concrete crack pixel-level segmentation

Dow, Hamish and Perry, Marcus and McAlorum, Jack and Pennada, Sanjeetha and Dobie, Gordon (2023) A novel directional lighting algorithm for concrete crack pixel-level segmentation. Proceedings of SPIE - The International Society for Optical Engineering, 12486. 124861L. ISSN 0277-786X (https://doi.org/10.1117/12.2657235)

[thumbnail of A_Novel_Directional_Lighting_Algorithm_for_Concrete_Crack_Pixel-level_Segmentation]
Preview
Text. Filename: A_Novel_Directional_Lighting_Algorithm_for_Concrete_Crack_Pixel-level_Segmentation.pdf
Final Published Version

Download (44MB)| Preview

Abstract

External lighting is required for autonomous inspections of concrete structures in low-light environments; however, previous studies have only considered uniformly diffused lighting to illuminate images. This study proposes a novel algorithm that utilises angled and directional lighting to obtain pixel-level segmentation of concrete cracks. The method applies a concrete crack detection algorithm to separate images, each illuminated with lighting from a different direction. Using a bitwise OR operation, the findings from all images are combined; the resulting image highlights the extremities of any present cracks in all lighting directions. When tested on a dataset of cracks ranging in widths from 0.07 mm to 0.3 mm, the algorithm obtained recall, precision and F1 score results of 77%, 84% and 92%, respectively. The algorithm was able to correctly segment cracks that were deemed too thin for similar diffused lighting segmentation methods found in literature. The proposed directional lighting algorithm has the potential to improve concrete inspections in low-light environments.

ORCID iDs

Dow, Hamish ORCID logoORCID: https://orcid.org/0000-0002-1431-7063, Perry, Marcus ORCID logoORCID: https://orcid.org/0000-0001-9173-8198, McAlorum, Jack ORCID logoORCID: https://orcid.org/0000-0001-8348-9945, Pennada, Sanjeetha and Dobie, Gordon ORCID logoORCID: https://orcid.org/0000-0003-3972-5917;