Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/76282
Title: Improved image unevenness reduction and thresholding methods for effective asphalt X-ray CT image segmentation
Authors: Chen, L 
Wang, YH 
Keywords: Asphalt pavements
Construction materials
Image techniques
Tomography
X-ray analysis
Issue Date: 2017
Publisher: American Society of Civil Engineers
Source: Journal of computing in civil engineering, 2017, v. 31, no. 4, 4017002 How to cite?
Journal: Journal of computing in civil engineering 
Abstract: The internal structure of asphalt mixture revealed by X-ray computed tomography (CT) images provides useful information in a variety of civil engineering applications. This paper studied two image processing issues commonly encountered in asphalt X-ray CT images: image intensity unevenness caused by beam hardening effect and unrealistic image segmentation. Inspired by nonuniform illumination correction techniques, a grayscale morphological method is proposed to solve image intensity unevenness. Two modified multilevel thresholding methods are developed to effectively divide the asphalt CT images into three phasesair voids, binder, and aggregates. The comparisons of existing multilevel thresholding methods with the ones developed in this study indicate that the proposed methods provide more consistent, robust, and accurate results. Moreover, the developed thresholding objective functions enable a user to easily adjust thresholds to match visual examination or laboratory test results through modifying a single parameter. The developed methods help improve the analysis of asphalt CT images for various engineering applications.
URI: http://hdl.handle.net/10397/76282
ISSN: 0887-3801
EISSN: 1943-5487
DOI: 10.1061/(ASCE)CP.1943-5487.0000631
Appears in Collections:Journal/Magazine Article

Access
View full-text via PolyU eLinks SFX Query
Show full item record

SCOPUSTM   
Citations

1
Citations as of Nov 13, 2018

Page view(s)

9
Last Week
1
Last month
Citations as of Nov 11, 2018

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.