基于岩土材料层析成像的孔隙网络构建算法

doi:10.19657/j.geoscience.1000-8527.2019.02.10

摘要/Abstract

摘要：

孔隙网络模型是研究多孔介质和其中颗粒的关键。现有的构建孔隙网络模型的算法有许多,但人们开始越来越多地研究基于图像的算法。将讨论一种基于中轴线的孔隙网络构建算法,并且对原有算法中孔隙中心表示方法不合理及缺少中轴线上各点坐标的问题提出改进,改进后的算法提出了孔隙中心的另一种表示方法,并且中轴线上任一点的坐标都被标出。在构建孔隙网络模型后利用该模型计算孔隙体积和喉径,并且针对过度分割的问题提出基于欧氏距离及基于分水岭算法的两种融合算法对孔隙和孔喉进行融合。

关键词: 孔隙网络, 孔喉, 岩土材料, 层析成像, 图像处理

Abstract:

Pore network model is important in researching for granular or porous media.Many methods have been proposed to construct a pore network, and researchers have put increasing efforts on the image-based methods. This paper presents a pore network construction method based on the medial axis method, as the original method cannot represent the pore center correctly and the coordinates of the axial points are unknown. The improved method proposed in this study represents the pore centers, and the coordinates of all axial points are accessible. After constructing the pore network, the pore volume and throat size can be calculated. Our work also presents an Euclidean distance-based and a watershed-based method to solve the over-segmentation problem.

Key words: pore network, throat, geomaterial, tomography, image process

中图分类号:

P642

任显卓, LINDEN Joost van der, NARSILIO Guillermo. 基于岩土材料层析成像的孔隙网络构建算法[J]. 现代地质, 2019, 33(02): 345-356.

REN Xianzhuo, LINDEN Joost van der, NARSILIO Guillermo. Developing a Pore Network Construction Algorithm for Computerized Tomography Images of Geomaterials[J]. Geoscience, 2019, 33(02): 345-356.

图/表 15

图1 孔隙网络构建算法流程图

Fig.1 Flowchart showing the steps of pore network construction

图2 标记规则 (a)被标记后的骨架;(b)相邻点示例

Fig.2 Rule of labelling

图3 骨架与图的关联 (a)骨架示例;(b)用图表示图(a)中的骨架

Fig.3 Relationship between skeleton and graph

图4 骨架分析流程图 (a)计算骨架数量流程图;(b)分割“ListofJunctionVoxels”链表流程图;(c)识别交汇处流程图;(d)识别边及其端点流程图

Fig.4 Flow charts of the skeleton analysis procedures

图5 距离变换操作生成的图像 (a)二维图像变换;(b)三维图像变换;左边为原图,右边是距离变换图

Fig.5 Images produced by the distance transforming algorithm

图6 距离变换算法生成的图像 (a)原始图像;(b)棋盘法生成的结果;(c)城市街区法生成的结果;(d)欧式距离法生成的结果

Fig.6 Images produced by the distance transforming algorithm

图7 喉径的计算结果 (a)调和平均值示例;(b)算法的输出示例

Fig.7 Results of the throat size calculation

图8 分水岭算法示例 (a)二维灰度图作为输入;(b)根据输入图绘制的地形表面图;(c)二维图像中的局部极小点和分水线;(d)水位升高过程

Fig.8 Process of the watershed algorithm

图9 分水岭算法分割结果 (a)原图像的分割结果;(b)经模糊处理后的分割结果

Fig.9 Results of the watershed segmentation

图10 根据融合后结果绘制的图像 (a)二维图像融合结果,使用图9(a)作为分水岭分割图;(b)二维图像融合结果,使用图9(b)作为分水岭分割图; (c)(d)三维图像融合结果,分别使用不同的分水岭分割图

Fig.10 Plot showing the results of watershed-based algorithm

图11 根据基于欧式距离的融合结果绘制的图像 (a)二维图像融合结果;(b)三维图像融合结果

Fig.11 Plot showing the results of the Euclidean distance-based algorithm

图12 球面交汇

Fig.12 Example of intersection of spherical surfaces

图13 根据改进后的基于欧氏距离的融合算法绘制的图像 (a)二维空间中的结果;(b)三维空间中的结果

Fig.13 Plots showing the results of the modified Euclidean distance-based merging algorithm

图14 根据改进后的基于分水岭算法的融合算法绘制的图像 (a)二维空间内的结果;(b)三维空间内的结果

Fig.14 Plot showing the results of the modified watershed-based merging algorithm

图15 结果分析 (a)未经融合的数据;(b)基于欧式距离的融合算法;(c)基于分水岭算法的融合结果;(d)改进后的基于欧式距离的融合结果;(e)改进后的基于分水岭算法的融合结果;图中喉径和孔隙半径的单位为像素

Fig.15 Result analyses

参考文献 23

[1]	SILVEIRA A. A method for determining the void size distribution curve for filter materials[M]// BRAUNS J,HEIBAUM M,SCHULER U. Filters in Geotechnical and Hydraulic Engineering.Karlsrule:International Conference Committee on Filters and Filtration Phenomena in Geotechnical Engineering, 1993: 71-74.
[2]	LOCKE M, INDRARATNA B. A new model for the behaviour of granular filters[M]// Faculty of Engineering and Information Sciences of University of Western Australia. Proceedings of Fourth Australia New Zealand Young Geotechnical Professionals Conference. Perth: University of Western Australia, 2000: 147.
[3]	SILVEIRA A, DE LORENA PEIXOTO T, NOGUEIRA J. On void size distribution of granular materials[M]// Committee of the 5th Pan-American Conference on Soil Mechanics and Foundation Engineering.Proceedings of the 5th Pan-American conference on Soil Mechanics and Foundation Engineering. Buenos Aires: Committee of the 5th Pan-American Conference on Soil Mechanics and Foundation Engineering, 1975: 161-177.
[4]	DONG H, TOUATI M, BLUNT M J. Pore network modeling: Analysis of pore size distribution of Arabian core samples[M]// SPE.Pore Network Modeling: Analysis of Pore Size Distribution of Arabian Core Samples. Richardson:Society of Petroleum Engineers, 2007: 1-5.
[5]	LINDQUIST W B. Investigating 3D geometry of porous media from high resolution images[J]. Physics and Chemistry of the Earth, Part A: Solid Earth and Geodesy, 1999,24(7):593-599. DOI URL
[6]	LINDQUIST W B, VENKATARANGAN A, DUNSMUIR J, et al. Pore and throat size distributions measured from synchrotron X-ray tomographic images of Fontainebleau sandstones[J]. Journal of Geophysical Research: Solid Earth, 2000,105(B9):21509-21527.
[7]	BILTREYS T, DE BOEVER W, CNUDDE V. Imaging and image-based fluid transport modeling at the pore scale in geological materials: A practical introduction to the current state-of-the-art[J]. Earth-Science Reviews, 2016,155:93-128. DOI URL
[8]	SILIN D, PATZEK T. Pore space morphology analysis using maximal inscribed spheres[J]. Physica A: Statistical Mechanics and Its Applications, 2016,371(2):336-360. DOI URL
[9]	TAYLOR H F, O’SULLIVAN C, SIM W W. A new method to identify void constrictions in micro-CT images of sand[J]. Computers and Geotechnics, 2015,69:279-290. DOI URL
[10]	LINDQUIST W B, LEE S M, COKER D A, et al. Medial axis analysis of void structure in three-dimensional tomographic images of porous media[J]. Journal of Geophysical Research: Solid Earth, 1996,101(B4):8297-8310.
[11]	HUNT A, EWING R, GHANBARIAN B. Percolation Theory for Flow in Porous Media[M]. New York:Springer, 2014: 1-466.
[12]	JONES A C, ARNS C H, SHEPPARD A P H, et al. Assessment of bone ingrowth into porous biomaterials using MICRO-CT[J]. Biomaterials, 2007,28(15):2491-2504. DOI URL PMID
[13]	MENDOZA F, VERBOVEN P, MEBATSION H K, et al. Three-dimensional pore space quantification of apple tissue using X-ray computed microtomography[J]. Planta, 2007,226(3):559-570. DOI URL PMID
[14]	PUDNEY C. Distance-ordered homotopic thinning: a skeletonization algorithm for three-dimensional digital images[J]. Computer Vision and Image Understanding, 1998,72(3):404-413. DOI URL
[15]	BORGEFORS G. On digital distance transforms in three dimensions[J]. Computer Vision and Image Understanding, 1996,64(3):368-376. DOI URL
[16]	SOILLE P, VINCENT L M. Determining watersheds in digital pictures via flooding simulations[J]. Journal of International Society for Optics and Photonics, 1990,1360:240-250.
[17]	LEE T C, KASHYAP R L, CHU C N. Building skeleton models via 3-D medial surface axis thinning algorithms[J]. Graphical Models and Image Processing, 1994,56(6):462-478. DOI URL
[18]	COEURJOLLY D, MONTANVERT A. Optimal separable algorithms to compute the reverse Euclidean distance transformation and discrete medial axis in arbitrary dimension[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007,29(3):437-448. URL PMID
[19]	BLUM H. A transformation for extracting new descriptors of shape[J]. Models for the Perception of Speech and Visual Form, 1967,19(5):362-380.
[20]	FATT L. The network model of porous media[J]. Transactions of the American Institute of Mining and Metallurgical Engineers, 1956,207(7):164-181.
[21]	CELIA M, REEVES P, FERRAND L. Recent advances in pore scale models for multiphase flow in porous media[J]. Reviews of Geophysics, 1995,33:1049-1057. DOI URL
[22]	DOUBE M, KŁOSOWSKI M M, ARGANDA-CARRERAS I, et al. BoneJ: free and extensible bone image analysis in Image[J][J]. Bone, 2010,47(6):1076-1079. URL PMID
[23]	ZAKIROV T R, GALEEV A A, KOROLEV E A, et al. Flow properties of sandstone and carbonate rocks by X-ray computed tomography[J]. Current Science, 2016,110(11):2142. DOI URL