Automatic Measurement of Concrete Crack Width in 2D Multiple-phase Images for Building Safety Evaluation

9Citations
Citations of this article
10Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Most of image processing techniques, that are developed for measuring concrete crack width in building safety evaluation process, have focused on two-phase piecewise constant images. However, the crack images in real applications are usually 2D multiple-phase piecewise constant images, which are defined as having more than 2 piecewise constant intensity phases. In this paper, we proposed a novel technique for automatically measuring crack widths in 2D multiple-phase images, which includes three main steps. Firstly, we used crack enhancement filter to enhance crack and remove other unintended objects in the intensity inhomogeneous image. Secondly, we utilized a B-spline level set model to automatically extract crack locations. Finally, we proposed to use a Savitzky-Golay filter based method to measure crack widths along the whole detected cracks. Several experiments are demonstrated to show the superior of our method in measuring crack width from images of real scene of damaged building.

Cite

CITATION STYLE

APA

Nguyen, H. N., Nguyen, T. Y., & Pham, D. L. (2018). Automatic Measurement of Concrete Crack Width in 2D Multiple-phase Images for Building Safety Evaluation. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10752 LNAI, pp. 638–648). Springer Verlag. https://doi.org/10.1007/978-3-319-75420-8_60

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free