Efficient pore network extraction method based on the distance transform

Abstract

Digital twins of materials allow to achieve accurate predictions that help creating novel and tailor-made materials with higher standards. In this paper, we are interested in the characterization of porous media. Our attention is drawn to develop a method to describe accurately the pore network microstructure of porous materials as presented in[7]. This work proposes an efficient algorithm based on the distance transform method[12] which is a widely used method in image processing. The followed approach suggests that a distance transform map, obtained from a microstructure image, passes through different steps. Starting from local maxima extraction and filtering operation, to end up with another distance transform with source propagation. We illustrate our algorithm with the well-known Pore Network Model of the literature[13], which supposes that the pore structure is either a network of connected cylinders or cylinders and spheres. Our approach is also applied on multi-scale Boolean random models modelling complex porous media microstructures[11]. The porous media morphological characteristics extracted could be used to simulate complex phenomena as the physisorption isotherms or other experimental techniques.