Generation of ground truth images to validate micro-CT image-processing pipelines

2018 ◽  
Vol 37 (6) ◽  
pp. 412-420 ◽  
Author(s):  
Steffen Berg ◽  
Nishank Saxena ◽  
Majeed Shaik ◽  
Chaitanya Pradhan
Keyword(s):  
Pore Space ◽  
Ground Truth ◽  
Fluid Distribution ◽  
Micro Ct ◽  
Pore Scale ◽  
Binary Images ◽  
Rock Heterogeneity ◽  
Wide Range ◽  
Machine Learning Approach ◽  
Subsurface Engineering

Digital rock technology and pore-scale physics have become increasingly relevant topics in a wide range of porous media with important applications in subsurface engineering. This technology relies heavily on images of pore space and pore-level fluid distribution determined by X-ray microcomputed tomography (micro-CT). Digital images of pore space (or pore-scale fluid distribution) are typically obtained as gray-level images that first need to be processed and segmented to obtain the binary images that uniquely represent rock and pore (including fluid phases). This processing step is not trivial. Rock complexity, image quality, noise, and other artifacts prohibit the use of a standard processing workflow. Instead, an array of strategies of increasing sophistication has been developed. Typical processing pipelines consist of filtering, segmentation, and postprocessing steps. For each step, various choices and different options exist. This makes selection and validation of an optimum processing pipeline difficult. Using Darcy-scale quantities as a benchmark is not a good option because of rock heterogeneity and different scales of observation. Here, we present a conceptual workflow where noisy images are derived from a ground truth by systematically including typical image artifacts and noise. Artifacts and noise are not simply added to the images. Instead, tomographic forward projection and reconstruction steps are used to incorporate the artifacts in a physically correct way. A proof of concept of this workflow is demonstrated by comparing seven different image-segmentation pipelines ranging from absolute thresholding to a machine-learning approach (Trainable Weka Segmentation). The Trainable Weka Segmentation showed the best performance of the tested methods.

scholarly journals 4D imaging of sub-second dynamics in pore-scale processes using real time synchrotron x-ray tomography

2016 ◽  
Author(s):  
Katherine J. Dobson ◽  
Sophia B. Coban ◽  
Sam A. McDonald ◽  
Joanna Walsh ◽  
Robert Atwood ◽  
...  
Keyword(s):  
Real Time ◽  
High Frequency ◽  
High Speed ◽  
Motion Blur ◽  
Transport Processes ◽  
Fluid Distribution ◽  
Pore Scale ◽  
X Ray ◽  
Wide Range ◽  
In Situ Experiments

Abstract. A variable volume flow cell has been integrated with state-of-the-art ultra-high speed synchrotron x-ray tomography imaging. The combination allows the first real time (sub-second) capture of dynamic pore (micron) scale fluid transport processes in 4D (3D + time). With 3D data volumes acquired at up to 20 Hz, we perform in situ experiments that capture high frequency pore-scale dynamics in 5–25 mm diameter samples with voxel (3D equivalent of a pixel) resolution of 2.5 to 3.8 µm. The data are free from motion artefacts, can be spatially registered or collected in the same orientation making them suitable for detailed quantitative analysis of the dynamic fluid distribution pathways and processes. The method presented here are capable of capturing a wide range of high frequency non equilibrium pore-scale processed including wetting, dilution, mixing and reaction phenomena, without sacrificing significant spatial resolution. As well as fast streaming (continuous acquisition) at 20 Hz, it also allows larger-scale and longer term experimental runs to be sampled intermittently at lower frequency (time-lapse imaging); benefiting from fast image acquisition rates to prevent motion blur in highly dynamic systems. This marks a major technical breakthrough for quantification of high frequency pore scale processes: processes that are critical for developing and validating more accurate multiscale flow models through spatially and temporally heterogeneous pore networks.

scholarly journals A Pore-Scale Simulation on Thermal-Hydromechanical Coupling Mechanism of Rock

Geofluids ◽  
2017 ◽  
Vol 2017 ◽  
pp. 1-12 ◽  
Author(s):  
Rui Song ◽  
Mengmeng Cui ◽  
Jianjun Liu ◽  
P. G. Ranjith ◽  
Yun Lei
Keyword(s):  
Pore Space ◽  
Water Flooding ◽  
Physical Parameters ◽  
Coupling Mechanism ◽  
Micro Ct ◽  
Pore Scale ◽  
Hydromechanical Coupling ◽  
Two Phase ◽  
Plastic Rock ◽  
Perfect Plastic

Thermal-hydromechanical (THM) coupling process is a key issue in geotechnical engineering emphasized by many scholars. Most existing studies are conducted at macroscale or mesoscale. This paper presents a pore-scale THM coupling study of the immiscible two-phase flow in the perfect-plastic rock. Assembled rock matrix and pore space models are reconstructed using micro-CT image. The rock deformation and fluid flow are simulated using ANSYS and CFX software, respectively, in which process the coupled physical parameters will be exchanged by ANSYS multiphysics platform at the end of each iteration. Effects of stress and temperature on the rock porosity, permeability, microstructure, and the displacing mechanism of water flooding process are analyzed and revealed.

scholarly journals 4-D imaging of sub-second dynamics in pore-scale processes using real-time synchrotron X-ray tomography

Solid Earth ◽  
2016 ◽  
Vol 7 (4) ◽  
pp. 1059-1073 ◽  
Author(s):  
Katherine J. Dobson ◽  
Sophia B. Coban ◽  
Samuel A. McDonald ◽  
Joanna N. Walsh ◽  
Robert C. Atwood ◽  
...  
Keyword(s):  
Real Time ◽  
High Frequency ◽  
High Speed ◽  
Motion Blur ◽  
Transport Processes ◽  
Fluid Distribution ◽  
Pore Scale ◽  
X Ray ◽  
Wide Range ◽  
In Situ Experiments

Abstract. A variable volume flow cell has been integrated with state-of-the-art ultra-high-speed synchrotron X-ray tomography imaging. The combination allows the first real-time (sub-second) capture of dynamic pore (micron)-scale fluid transport processes in 4-D (3-D + time). With 3-D data volumes acquired at up to 20 Hz, we perform in situ experiments that capture high-frequency pore-scale dynamics in 5–25 mm diameter samples with voxel (3-D equivalent of a pixel) resolutions of 2.5 to 3.8 µm. The data are free from motion artefacts and can be spatially registered or collected in the same orientation, making them suitable for detailed quantitative analysis of the dynamic fluid distribution pathways and processes. The methods presented here are capable of capturing a wide range of high-frequency nonequilibrium pore-scale processes including wetting, dilution, mixing, and reaction phenomena, without sacrificing significant spatial resolution. As well as fast streaming (continuous acquisition) at 20 Hz, they also allow larger-scale and longer-term experimental runs to be sampled intermittently at lower frequency (time-lapse imaging), benefiting from fast image acquisition rates to prevent motion blur in highly dynamic systems. This marks a major technical breakthrough for quantification of high-frequency pore-scale processes: processes that are critical for developing and validating more accurate multiscale flow models through spatially and temporally heterogeneous pore networks.

scholarly journals Pore-scale imaging and analysis of low salinity waterflooding in a heterogeneous carbonate rock at reservoir conditions

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Ahmed M. Selem ◽  
Nicolas Agenet ◽  
Ying Gao ◽  
Ali Q. Raeini ◽  
Martin J. Blunt ◽  
...  
Keyword(s):  
Oil Recovery ◽  
Carbonate Rock ◽  
Pore Space ◽  
Contact Angles ◽  
Fluid Distribution ◽  
Pore Scale ◽  
Low Salinity ◽  
Low Salinity Waterflooding ◽  
Capillary Pressures ◽  
Micro Tomography

AbstractX-ray micro-tomography combined with a high-pressure high-temperature flow apparatus and advanced image analysis techniques were used to image and study fluid distribution, wetting states and oil recovery during low salinity waterflooding (LSW) in a complex carbonate rock at subsurface conditions. The sample, aged with crude oil, was flooded with low salinity brine with a series of increasing flow rates, eventually recovering 85% of the oil initially in place in the resolved porosity. The pore and throat occupancy analysis revealed a change in fluid distribution in the pore space for different injection rates. Low salinity brine initially invaded large pores, consistent with displacement in an oil-wet rock. However, as more brine was injected, a redistribution of fluids was observed; smaller pores and throats were invaded by brine and the displaced oil moved into larger pore elements. Furthermore, in situ contact angles and curvatures of oil–brine interfaces were measured to characterize wettability changes within the pore space and calculate capillary pressure. Contact angles, mean curvatures and capillary pressures all showed a shift from weakly oil-wet towards a mixed-wet state as more pore volumes of low salinity brine were injected into the sample. Overall, this study establishes a methodology to characterize and quantify wettability changes at the pore scale which appears to be the dominant mechanism for oil recovery by LSW.

scholarly journals New Method of Oil Reservoir Rock Heterogeneity Quantitative Estimation from X-ray MCT Data

Energies ◽  
2021 ◽  
Vol 14 (16) ◽  
pp. 5103
Author(s):  
Irena Viktorovna Yazynina ◽  
Evgeny Vladimirovich Shelyago ◽  
Andrey Andreevich Abrosimov ◽  
Vladimir Stanislavovich Yakushev
Keyword(s):  
Quantitative Estimation ◽  
Pore Space ◽  
Core Sample ◽  
New Method ◽  
Reservoir Rock ◽  
Core Material ◽  
Oil Reservoir ◽  
Pore Scale ◽  
X Ray ◽  
Rock Heterogeneity

This paper considers a new method for “pore scale” oil reservoir rock quantitative estimation. The method is based on core sample X-ray tomography data analysis and can be directly used to both classify rocks by heterogeneity and assess representativeness of the core material collection. The proposed heterogeneity criteria consider the heterogeneity of pore size and heterogeneity of pore arrangement in the sample void and can thus be related to the drainage effectiveness. The classification of rocks by heterogeneity at the pore scale is also proposed when choosing a reservoir engineering method and may help us to find formations that are similar at pore scale. We analyzed a set of reservoir rocks of different lithologies using the new method that considers only tomographic images and clearly distributes samples over the structure of their pore space.

scholarly journals Moment-to-Moment Continuous Attention Fluctuation Monitoring through Consumer-Grade EEG Device

Sensors ◽  
2021 ◽  
Vol 21 (10) ◽  
pp. 3419
Author(s):  
Shan Zhang ◽  
Zihan Yan ◽  
Shardul Sapkota ◽  
Shengdong Zhao ◽  
Wei Tsang Ooi
Keyword(s):  
Ground Truth ◽  
Detection Accuracy ◽  
Continuous Performance Task ◽  
Eeg Signals ◽  
Gradual Onset ◽  
Machine Learning Approach ◽  
A New Technique ◽  
Thought Probes ◽  
Labeling Method ◽  
Fluctuation Measurement

While numerous studies have explored using various sensing techniques to measure attention states, moment-to-moment attention fluctuation measurement is unavailable. To bridge this gap, we applied a novel paradigm in psychology, the gradual-onset continuous performance task (gradCPT), to collect the ground truth of attention states. GradCPT allows for the precise labeling of attention fluctuation on an 800 ms time scale. We then developed a new technique for measuring continuous attention fluctuation, based on a machine learning approach that uses the spectral properties of EEG signals as the main features. We demonstrated that, even using a consumer grade EEG device, the detection accuracy of moment-to-moment attention fluctuations was 73.49%. Next, we empirically validated our technique in a video learning scenario and found that our technique match with the classification obtained through thought probes, with an average F1 score of 0.77. Our results suggest the effectiveness of using gradCPT as a ground truth labeling method and the feasibility of using consumer-grade EEG devices for continuous attention fluctuation detection.

scholarly journals Assessing the Relation between Mud Components and Rheology for Loss Circulation Prevention Using Polymeric Gels: A Machine Learning Approach

Energies ◽  
2021 ◽  
Vol 14 (5) ◽  
pp. 1377
Author(s):  
Musaab I. Magzoub ◽  
Raj Kiran ◽  
Saeed Salehi ◽  
Ibnelwaleed A. Hussein ◽  
Mustafa S. Nasser
Keyword(s):  
Machine Learning ◽  
Rheological Properties ◽  
Nearest Neighbor ◽  
Drilling Fluid ◽  
Gradient Boosting ◽  
K Nearest Neighbor ◽  
Wide Range ◽  
Machine Learning Approach ◽  
Drilling Operations

The traditional way to mitigate loss circulation in drilling operations is to use preventative and curative materials. However, it is difficult to quantify the amount of materials from every possible combination to produce customized rheological properties. In this study, machine learning (ML) is used to develop a framework to identify material composition for loss circulation applications based on the desired rheological characteristics. The relation between the rheological properties and the mud components for polyacrylamide/polyethyleneimine (PAM/PEI)-based mud is assessed experimentally. Four different ML algorithms were implemented to model the rheological data for various mud components at different concentrations and testing conditions. These four algorithms include (a) k-Nearest Neighbor, (b) Random Forest, (c) Gradient Boosting, and (d) AdaBoosting. The Gradient Boosting model showed the highest accuracy (91 and 74% for plastic and apparent viscosity, respectively), which can be further used for hydraulic calculations. Overall, the experimental study presented in this paper, together with the proposed ML-based framework, adds valuable information to the design of PAM/PEI-based mud. The ML models allowed a wide range of rheology assessments for various drilling fluid formulations with a mean accuracy of up to 91%. The case study has shown that with the appropriate combination of materials, reasonable rheological properties could be achieved to prevent loss circulation by managing the equivalent circulating density (ECD).

scholarly journals Determination of macro-scale soil properties from pore-scale structures: model derivation

2018 ◽  
Vol 474 (2209) ◽  
pp. 20170141 ◽  
Author(s):  
K. R. Daly ◽  
T. Roose
Keyword(s):  
Pore Space ◽  
Solid Particles ◽  
Underlying Structure ◽  
Pore Scale ◽  
Homogenization Procedure ◽  
Macro Scale ◽  
Model Derivation ◽  
Scale Structures ◽  
Homogenization Scheme

In this paper, we use homogenization to derive a set of macro-scale poro-elastic equations for soils composed of rigid solid particles, air-filled pore space and a poro-elastic mixed phase. We consider the derivation in the limit of large deformation and show that by solving representative problems on the micro-scale we can parametrize the macro-scale equations. To validate the homogenization procedure, we compare the predictions of the homogenized equations with those of the full equations for a range of different geometries and material properties. We show that the results differ by ≲ 2 % for all cases considered. The success of the homogenization scheme means that it can be used to determine the macro-scale poro-elastic properties of soils from the underlying structure. Hence, it will prove a valuable tool in both characterization and optimization.

scholarly journals Classification of unlabeled online media

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Sakthi Kumar Arul Prakash ◽  
Conrad Tucker
Keyword(s):  
Social Media ◽  
Real World ◽  
Graphical Model ◽  
Ground Truth ◽  
Classification Problem ◽  
Machine Learning Algorithms ◽  
Social Media Networks ◽  
Online Social Media ◽  
Wide Range

AbstractThis work investigates the ability to classify misinformation in online social media networks in a manner that avoids the need for ground truth labels. Rather than approach the classification problem as a task for humans or machine learning algorithms, this work leverages user–user and user–media (i.e.,media likes) interactions to infer the type of information (fake vs. authentic) being spread, without needing to know the actual details of the information itself. To study the inception and evolution of user–user and user–media interactions over time, we create an experimental platform that mimics the functionality of real-world social media networks. We develop a graphical model that considers the evolution of this network topology to model the uncertainty (entropy) propagation when fake and authentic media disseminates across the network. The creation of a real-world social media network enables a wide range of hypotheses to be tested pertaining to users, their interactions with other users, and with media content. The discovery that the entropy of user–user and user–media interactions approximate fake and authentic media likes, enables us to classify fake media in an unsupervised learning manner.

Sign in / Sign up

Export Citation Format

Share Document