Memory-effective methods and algorithms of shader visualization of digital core material model

Abstract

The paper deals with the task of real-time visualization of digital core material model (DCM) obtained by X-ray computed tomography. Novel approach, effective methods and algorithms for visualization of large allocated volume of DCM are proposed, which are based on high-performance programmable tessellation capability of commodity graphics cards with a multicore graphics processing unit (GPU). The methods provide an effective video memory usage and a high visualization rate for the ranges of absorption coefficient, corresponding to pore space or mineral skeleton of the core material. The solution proposed is based on visualization of trihedral polygonal models of visible voxels and on the original system for constructing such models on GPU by means of tessellation shaders developed. Based on these methods and algorithms, a program module is developed, which implements 3D visualization of the allocated volume of DCM and the construction of porosity plot. The program module was tested on the DCMs of Bazhenov formation and sandstone, that confirmed adequateness of developed solution to the task arisen.