Reconstructing granular particles from X-ray computed tomography using the TWS machine learning tool and the level set method

X-ray computed tomography (CT) has emerged as the most prevalent technique to obtain three-dimensional morphological information of granular geomaterials. A key challenge in using the X-ray CT technique is to faithfully reconstruct particle morphology based on the discretized pixel information of CT images. In this work, a novel framework based on the machine learning technique and the level set method is proposed to segment CT images and reconstruct particles of granular geomaterials. Within this framework, a feature-based machine learning technique termed Trainable Weka Segmentation is utilized for CT image segmentation, i.e., to classify material phases and to segregate particles in contact. This is a fundamentally different approach in that it predicts segmentation results based on a trained classifier model that implicitly includes image features and regression functions. Subsequently, an edge-based level set method is applied to approach an accurate characterization of the particle shape. The proposed framework is applied to reconstruct three-dimensional realistic particle shapes of the Mojave Mars Simulant. Quantitative accuracy analysis shows that the proposed framework exhibits superior performance over the conventional watershed-based method in terms of both the pixel-based classification accuracy and the particle-based segmentation accuracy. Using the reconstructed realistic particles, the particle-size distribution is obtained and validated against experiment sieve analysis. Quantitative morphology analysis is also performed, showing promising potentials of the proposed framework in characterizing granular geomaterials.

     

考虑颗粒棱角影响的直剪试验的离散元模拟

为了研究颗粒棱角对颗粒材料力学行为的影响,建立了具有不同棱角度的对称多面体颗粒,采用了一种简单并适合任意颗粒形状的接触本构模型,对三维离散元开源程序YADE进行了修改,研究了颗粒棱角度在模拟直剪试验中的影响以及接触力各向异性在剪切过程中的演化规律。研究结果表明,颗粒棱角度越小,颗粒间相互咬合自锁的作用越小,颗粒受剪更易转动,致使颗粒体系的剪切强度和剪胀性下降;竖向加载力越大,颗粒棱角度的影响越明显;法向接触力的各向异性在剪切过程中表现为先增后减最后趋向稳定的趋势;法向接触力的各向异性变化程度随颗粒棱角度的增大而增大。     

用于颗粒土微观力学行为试验的微型三轴试验仪

颗粒土的微观力学行为（如土颗粒的运动与破碎等）决定着其宏观应力-应变行为,如应变局部化、应力硬化等。为研究颗粒土的微观力学行为,开发了一台微型三轴试验装置。它的轴向加载系统由伺服控制的步进转动马达与涡轮传动的减速器组成,围压由GDS压力控制器提供,压力室采用高透光率,高强度的轻质材料制成。借助于X射线显微CT及图像处理分析技术,该装置能实现对干砂土微尺寸试样（直径为8 mm,高度为16 mm）在三轴剪切条件下微观特性的无损检测。采用该三轴试验装置对粒径为0.60～1.18 mm 的LBS（Leighton Buzzard sand）试样在1.5 MPa的围压下进行了试验。结果显示,试验装置测得应力-应变曲线合理,显微CT 图像特征清晰,能够用于颗粒土体微观土力学行为的试验研究。     

Evaluation of image analysis methods used for quantification of particle angularity

Angularity is an important parameter in the characterization of particle morphology that is used to interpret the transport history of particles in sedimentary deposits. In the past, visual classification using silhouette charts was widely used to determine particle angularity, but this approach is subjective and time‐consuming. With advances in modern image analysis techniques and low‐cost software packages, it is possible to rapidly quantify particle angularity more objectively than using visual classification methods. This study re‐examines the performance of three existing image analysis methods and one new image analysis procedure, applied to six rock and sediment samples that were visually different in angularity. To facilitate comparison between the angularity results, measurements were reduced to rankings for each aggregate sample. These results show that the four image analysis methods rank the angularity of the samples differently, and that none rank the mean angularity index in the same order as the angularity ranking using visual classification. Therefore, further research is needed to develop an image analysis method that can quantify the angularity of sedimentary particles more precisely.     

Notes on the effect of grain crushing on the granular soil behaviour

Two granular materials — alluvial quartzy Zbraslav sand and granular silica gel — were tested with the intention to demonstrate the effect of grain crushing. Stepwise transformation of the compression curve produced by progressive grain crushing was observed. Due to grain crushing, shear strength envelope became nonlinear, and the behaviour was no more physically isomorphous. The shear stress–strain diagrams acquire a typical wavy (garland-like) form, induced by periodic softening and hardening of the soil response. The intensity of grain crushing depends (in addition to stress level, grain resistance and time) on the shear path. In the crushing phase, initial porosity and angularity play a secondary role. Many other behavioural features common with granular soils (like increase in dilatancy with density and grain size) are suppressed. Grain crushing thus produces a qualitatively different feature of geomaterial behaviour with grave practical consequences (dense sand, e.g. starts to behave like loose).     

Micromechanical investigation of the particle size effect on the shear strength of uncrushable granular materials

Particle size strongly influences the shear strength of granular materials. However, previous studies of the particle size effect have focused mainly on the macroscopic behavior of granular materials, neglecting the associated micro-mechanism. In this study, the effect of particle size on the shear strength of uncrushable granular materials in biaxial testing is investigated using the discrete element method (DEM). First, a comprehensive calibration against experimental results is conducted to obtain the DEM parameters for two types of quartz sand. Then, a series of biaxial tests are simulated on sands with parallel particle size distributions to investigate the effect of particle size on macro- and microscopic behaviors. Finally, by adopting the rolling resistance method and the clump method, irregular-shaped particles are simulated to investigate how the particle size effect will be influenced by the particle shape. Simulation results demonstrate that (1) the peak shear strength increases with particle size, whereas the residual shear strength is independent of particle size; (2) the thickness of the shear band increases with the particle size, but its ratio decreases with particle size; (3) the particle size effect can be explained by the increase of friction utilization ratio with particle size; and (4) the particle size effect is more significant in granular materials that consist of particles with higher angularity.

     

Geotechnical properties of 3D-printed transparent granular soil

This paper proposes a 3D-printed transparent granular soil technique based on the contour rotation interpolation method, 3D printing and transparent soil technologies. Laboratory tests, including one-dimensional compression, direct shear and triaxial compression tests, are assessed for mechanical properties of 3D-printed transparent granular soil. The results show that 3D-printed transparent granular soil can be used to consider the effect of shape on macromechanical soil properties, which is an advantage that cannot be achieved by the previously presented transparent soil techniques. Subsequently, a simple model test of rigid flat plate penetration into transparent granular soil is performed. The obtained soil displacements using the PIV technique are compared with the classical shallow strain path method solution. This comparison indicates that the proposed 3D-printed transparent granular soil can capture the soil deformation pattern, although the accuracy needs to be further improved. The proposed 3D-printed transparent granular soil technique could be used for model test that needs to consider the shape of sand particle.

     

Evaluation of hydraulic conductivity of gap-graded granular soils based on equivalent void ratio concept

Gap-graded granular soils are used as construction materials worldwide, and their hydraulic conductivity depends on their relative content of coarse and fine grains, initial conditions, and particle shape. In this study, a series of constant head hydraulic conductivity tests were performed on gap-graded granular soils with different initial relative densities, fine contents, and particle shapes. The test results show that the hydraulic conductivity decreases with an increase in fine fraction and then remains approximately constant beyond the “transitional fine content.” The role of the structural effect on the hydraulic conductivity is different from that on the mechanical properties (such as stiffness and shear strength). This can be attributed to the degree of filling within inter-aggregate voids, disturbance of soil structure, and densified fine bridges between coarse aggregates. The equivalent void ratio concept was introduced into the Kozeny–Carman formula to capture the effect of fines (aggregates) on the “coarse-dominated” (“fine-dominated”) structure, and a simple model is proposed to capture the change of hydraulic conductivity of gap-granular soils. The model incorporates a structural variable to capture the effect of fines on “coarse-dominated” structure and coarse aggregates on “fine-dominated” structure. The performance of the model was verified with experimental data from this study and previously reported data compiled from the literature. The results reveal that the proposed model is simple yet effective at capturing the hydraulic conductivity of gap-graded granular soils with a wide range of fine contents, initial conditions, and particle shapes.

     

一种微结构颗粒体分布特征的分析方法

在扫描电镜法（SEM）图像二值化基础上,建立了以颗粒体面积为变量的面积累计曲线表示方法,进而建立了该曲线的曲率系数和颗粒体面积分布特点的不均匀系数。基于瓜州砂质粉土和天津黏土的SEM图像,分析了不同面积比对应的颗粒体特征。研究表明,砂质粉土和黏土的颗粒体面积累计曲线、曲率系数和不均匀系数具有明显的差异性。砂质粉土的颗粒体面积累计曲线是一条平滑的曲线,表现为不良级配土的粒径分布特点;而黏土的颗粒体面积累计曲线具有明显的台阶,表现为某粒组缺失的粒径分布特点。另外,砂质粉土的颗粒体面积累计曲线曲率系数和不均匀系数均明显大于黏土的对应系数。该分析方法可以用来量化不同土的SEM照片之间的差异,从而为岩土微结构的定量化研究提供了一条新的思路。     

Investigations on Influence of Bio-Geo Interface on Suction Characteristics of Fine-Grained Soil

Dependence of soil suction on factors such as formation of the soil, its particle size, mineralogy, pore-size distribution characteristics, water content, fabric, stress history and presence of salts has been extensively investigated by researchers. However, the influence of the microbial presence and its activity on the soil suction has not yet received researchers’ attention. With this in view, a methodology that facilitates investigation regarding the presence of ‘soil-microbial interface’, which has been termed as ‘bio-geo interface’, has been proposed in this study. Furthermore, the influence of this ‘interface’ on suction characteristics of such soils has also been investigated and the utility of the proposed methodology in addressing such issues, successfully, has been demonstrated.     

离散单元法对粒状土的微观特性研究探讨

粒状土的微观结构和微观力学被认为是其宏观力学和体积特性的内在根本因素,近年来得到越来越多的关注和研究。离散单元法作为一种研究颗粒材料的数值模拟计算方法,比试验方法快捷、简便、经济,而且能够容易得到在实验室试验中很难或无法得到的更多重要的微观结构和微观力学的信息,近年来得到越来越多应用。本文介绍了离散单元法对土的微观特性研究的一些最新方法和进展,对数值建模中的一些重要方面如比重(质量)放大、树脂薄膜模拟等方面进行了阐述,对离散单元法在土的微观结构分析(如颗粒旋转、颗粒位移、中尺度孔隙率分布)的一些最新研究作了分析和介绍。分析表明,离散单元法是研究粒状土的微观特性的一个有力工具,可以对土的宏观特性从微观角度得到更好的解释和认识。     

On the evaluation of internal stability of gap-graded soil: a status quo review

Internal instability is a phenomenon of fine particle redistribution in granular materials under the seepage action and consequent change in the soil’s internal structure and hydraulic and mechanical properties. It is one of the primary causes of failures of sand-gravel foundations and embankment dams. The criteria establishment is considered the key to solving the erosion problems, so the existing internal stability criteria need a review and classification to study the recent development trends in soil seepage and erosion. Therefore, this paper aims at reviewing the internal stability factors of gap-graded soil with a focus on the internal erosion mechanism and internal stability evaluation based on geometric and hydraulic criteria. Firstly, the paper compared the effect of several commonly used geometric criteria for gap-graded soil evaluation, such as particle size, fine content, void ratio, and fractal dimension. Furthermore, it provided a hydraulic criteria overview and analyzed the effects of the hydraulic gradient, hydraulic shear stress, confining pressure, and pore velocity on internal erosion. The geometric–hydraulic coupling methods were introduced, with a detailed elaboration of the erosion resistance index method based on accumulated dissipated energy. The capabilities and limitations of these criteria were discussed throughout the paper. It was found that combined Kezdi’s criterion and Kenney and Lau’s criterion is more reliable to evaluate internal stability of soil. The gap-graded soil with fine particle content higher than 35% is not necessarily internally stable. Finally, the energy-based method (erosion resistance index method) can effectively reproduce the total amount of erosion mass and the final spatial distribution of fine particles and identifies erosion. The review's outcome can be used as a basis to evaluate the internal erosion risk for gap-graded soils. The evaluation methods discussed here can help identify the zones of relatively high erosion potential.

     

K0条件下颗粒形状特征对砂土剪切模量的影响

为了研究颗粒形状特征对砂土剪切模量的影响,选取4种形状不同的砂样（福建标准砂和人工石英砂）作为研究对象,通过扫描电镜获取其颗粒微观图像,使用Image Pro Plus软件和自编Matlab程序提取试样颗粒的形状特征。采用改进的半径角和轮廓指数综合描述颗粒偏离圆或椭圆的程度以及棱角变化程度,并基于这两个参数提出颗粒形状因子来综合表征颗粒的形状特征。之后在K0条件下通过弯曲元试验获得不同竖向荷载对应的4个试样的剪切模量,试验结果表明,砂土颗粒的形状因子与砂土的剪切模量具有良好的相关性。最后基于Hertz-Mindlin接触模型,推导出引入形状因子的砂土剪切模量理论计算公式,该计算公式与实测值吻合良好。     

Vermiculate artefacts in image analysis of granular materials

Some reported analyses of images of deforming granular materials have generated surprising vermiculate strain features which are difficult to reconcile with the mechanics of deformation of granular matter. Detailed investigation using synthetic images and improved processing of images of laboratory experiments indicates that such features can emerge as a consequence of the image acquisition (sensor, contrast, resolution), the subsequent image correlation implementation, and the user’s choice of processing parameters. The two principal factors are: (i) the texture and resolution of the images and (ii) the algorithm used to achieve sub-pixel displacement resolution. Analysis of the images using a sub-pixel interpolation algorithm that is more robust than that used originally eliminates the vermiculate features for images with moderate resolution and texture. However, erroneous features persist in images with low resolution and poor texture. Guidance is provided on ways in which such artefacts can be avoided through improved experimental and image analysis techniques.     

Characterisation of conduction phenomena in soils at the particle-scale: Finite element analyses in conjunction with synthetic 3D imaging

This paper describes an innovative method to characterise conduction parameters in geomaterials at the particle-scale. The technique is exemplified using 3D synthetic grain packing generated with discrete element approaches. This creates a geo-mechanically viable user-defined 3D granular image through which the particle skeleton and the corresponding pore network are constructed. Images are then imported into the finite element analyses to solve the governing equations of hydraulic and thermal conduction. Navier–Stokes equation is uniquely upscaled to Darcy’s law to assess hydraulic conductivity in soils, while a similar approach implements the Fourier equation to evaluate thermal conduction through grain chains and pore network. High performance computing is used to meet demanding numerical calculations of 3D meshed geometries. Packing density (i.e., porosity) and inter-particle contact areas are explored as variables to highlight the effects of pore volume and inter-particle contact condition in hydraulic and thermal conduction. This emerging technique allows not only characterising the macro-scale behaviour of conduction phenomena in soils but also quantifying and visualising the preferential and local conduction behaviour at the particle-scale. Laboratory measurements of hydraulic and thermal conductivities support numerically obtained results and validate the viability of the new methods used herein. This study introduces an alternative way to determine physical parameters of soils using emerging technology of rigorous numerical simulations in conjunction with 3D images, and to enable fundamental observation of particle-scale mechanisms of macro-scale manifestation.     

Micromorphology and mineralogy of gullied soils in Anambra State,Nigeria

Detailed studies on the micromorphology and mineralogy of two guillied soils in Anambra State of Nigeria were carried out. Thin sections were prepared and point counting technique used in investigating the pedological features. Micromorphological analysis indicates that the S-matrix of the soil have mainly sepic and inundulic with few asepic plasmic fabric. Distribution patterns are agglomeroplasmic to porphyroskelic and granular.Clay mineralogy was determined by x-ray diffractometry and the result reveals that the soils are mostly dominated by kaolinite and quartz.Macromorphologically, the soils are very deep, well drained and coarse to medium textured with granular and medium subangular structures.The occurrence of ferriargillans in the B-horizon is a strong indication of pedotranslocation. Other contemporary processes or regrouping phenomena are pedotranslocation, pedoturbation, pedocompaction and pedotransformation of minerals. The role of clay mineralogy type in soil erosion problems is discussed.     

A tomographic imagery segmentation methodology for three-phase geomaterials based on simultaneous region growing

X-ray computed tomography is a powerful non-destructive technique used in many domains to obtain the three-dimensional representation of objects, starting from the reconstitution of two-dimensional images of radiographic scanning. This technique is now able to analyze objects within a few micron resolutions. Consequently, X-ray microcomputed tomography opens perspectives for the analysis of the fabric of multiphase geomaterials such as soils, concretes, rocks and ceramics. To be able to characterize the spatial distribution of the different phases in such complex and disordered materials, automated phase recognition has to be implemented through image segmentation. A crucial difficulty in segmenting images lies in the presence of noise in the obtained tomographic representation, making it difficult to assign a specific phase to each voxel of the image. In the present study, simultaneous region growing is used to reconstitute the three-dimensional segmented image of granular materials. First, based on a set of expected phases in the image, regions where specific phases are sure to be present are identified, leaving uncertain regions of the image unidentified. Subsequently, the identified regions are grown until growing phases meet each other with vanishing unidentified regions. The method requires a limited number of manual parameters that are easily determined. The developed method is illustrated based on three applications on granular materials, comparing the phase volume fractions obtained by segmentation with macroscopic data. It is demonstrated that the algorithm rapidly converges and fills the image after a few iterations.     

Fractal analysis of the roughness and size distribution of granular materials

The roughness, i.e. general shape and surface irregularity, of particulate soil is an important characteristic that affects the mass behavior of the soil. Characterization of roughness has typically been limited to visual comparison of particles to standard charts, although other more quantitative methods such as Fourier analysis have also been used. Particle size distribution is another important mass-behavioral characteristic of granular soils, and similar to roughness, is defined within limited boundaries. Fractal geometry can be applied to irregular or fragmented patterns such as roughness and grain size distribution to provide quantifying and unique numerical values. This paper presents an evaluation of the applicability of fractal dimensioning techniques to the quantification of both physical particle roughness and grain size distribution of granular soil. The divider and the area-perimeter fractal dimensioning techniques are used to quantify roughness of planar profiles of individual sand grains. The characterization of the size distribution of granular material using fractal geometry is evaluated through Korcak's fragmentation theory. As shown herein, both the divider and the area-perimeter fractal dimensioning techniques are useful in characterizing soil particle roughness, and the results confirm the importance of differentiating between textural and structural aspects of roughness. Fractal geometry can also be used to quantify the size distribution of granular soils with relatively well-graded size distributions.