Deep Learning of Multiresolution X-Ray Micro-Computed-Tomography Images for Multiscale Modeling
Samuel J. Jackson, Yufu Niu, Sojwal Manoorkar, Peyman Mostaghimi, Ryan T. Armstrong
Abstract
Field-of-view and resolution trade-offs in x-ray micro-computed-tomography (micro-CT) imaging limit the characterization, analysis, and model development of multiscale porous systems. To this end, we develop an applied methodology utilizing deep learning to enhance low-resolution (LR) images over large sample sizes and create multiscale models capable of accurately simulating experimental fluid dynamics from the pore (microns) to continuum (centimeters) scale. We develop a three-dimensional (3D) enhanced deep-superresolution (EDSR) convolutional neural network to create superresolution (SR) images from LR images, which alleviates common micro-CT hardware and/or reconstruction defects in high-resolution (HR) images. When paired with pore-network simulations and parallel computation, we can create large 3D continuum-scale models with spatially varying flow and material properties. We quantitatively validate the workflow at various scales using direct HR and SR image similarity, pore-scale material and/or flow simulations, and continuum-scale multiphase-flow experiments (drainage-immiscible flow pressures and 3D fluid-volume fractions). The SR images and models are comparable to the HR ground truth and generally accurate to within experimental uncertainty at the continuum scale across a range of flow rates. They are found to be significantly more accurate than their LR counterparts, especially in cases where a wide distribution of pore sizes are encountered. The applied methodology opens up the possibility to image, model, and analyze truly multiscale heterogeneous systems that are otherwise intractable.