Global image segmentation for two-phase snd multiphase geomaterial characterization
Date
2017
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
The effectiveness of five global thresholding techniques, to accurately segment different two-phase geomaterials (solids and air), was evaluated in this work. X-ray computed tomography (CT) images taken from pervious concrete, glass bead, and silica sand specimens were analyzed for evaluating the five chosen methods. The core algorithm of each of these methods was coded as a standalone graphical user interface (GUI) application software. From the results, it can be said that, no single image segmentation technique performs well over a wide range of material and that the performance of each image segmentation technique varies depending on the type and state of the analyzed media. ☐ X-ray CT images of three-phase (solids, water, and air) silica sand and glass bead specimens were analyzed and used for evaluating the segmentation performances of three methods. Based on the observations made on these methods, a refined statistical-based global segmentation algorithm is proposed for segmenting partially saturated granular geomaterials. The findings for the silica sand specimen showed that the proposed technique estimated void ratio with 1.52 percent error and degree of saturation with 4.35 percent error. On the other hand, void ratio for the glass bead specimen yielded a high percent error of 15.63 whereas degree of saturation had a very low percent error of 0.34. By comparing the results of the proposed technique with the other Otsu-based techniques, it is concluded that the proposed algorithm performs better in segmenting three-phase granular geomaterials.
Description
Keywords
Applied sciences