Important announcement of COVID 19 - Click here

Webinar: Grayscale representative elementary volumes: An innovative approach to investigate pore-scale REVs from raw micro-CT images

Event Type

Event Date

Thursday, October 1, 2020

Event Location

Event Address


Event Start

1200 (AWST)

Event End

1300 (AWST)

Event Details

Title and Summary:

Grayscale representative elementary volumes: An innovative approach to investigate pore-scale REVs from raw micro-CT images

Representative Elementary Volumes (REVs) are at the foundation of measuring rock properties that capture local heterogeneities of the rock structure at a particular length-scale for upscaling purposes. High-resolution micro-computed tomography (micro-CT) images of rocks have allowed a full 3D characterization of rock structures at pore-scale. These micro-CT images store information about rock structure as variations in the gray-level intensities or CT numbers. However, the direct use of these information-rich raw micro-CT images for rock characterization has not been possible due to a limited number of rock properties that can be calculated from them. In this study, we implement a novel texture characterization technique called the Gray-level Size Zone Matrix (GLSZM) to analyze the raw micro-CT images. We apply the GLSZM approach to homogeneous and heterogeneous sandstones and carbonates and show that this method highlights important rock features such as mineralogical heterogeneities and sub-resolution porosity. Considering these features, we calculate GLSZM statistics, that serve as proxies to porosity and permeability, which are crucial petrophysical properties. Comparing the trends of these proxies to petrophysical properties at various scales and spatial locations of the rock sample, we then infer Grayscale REVs (GREVs) and validate it using existing literature. Finally, we show that using the GLSZM-based approach, we can infer GREVs in a robust, reproducible, and fast manner. These GREVs can then serve as a priori for further petrophysical characterization of rock samples. 


Ankita Singh is a Ph.D. student at the School of Minerals and Energy Resources Engineering at UNSW, Sydney. Her work focusses on implementing texture analysis techniques for rock characterization by directly using raw x-ray images. Her Ph.D. work has been published in reputed journals such as Water Resources Research and Geophysical Research Letters. She also won the 'Best Engineering/Environmental Student Paper' at AEGC 2019 in Perth and was the 2019 Finalist at the UNSW Three Minute Thesis Competition. 

Register now: