Three-dimensional fabric orientation visualisation technique for distributed fractures using X-ray computed tomography

Abstract X-ray computed tomography is an extremely useful tool for non-destructive inspection of material internal damage. However, how to quantitatively portray the distributions and orientations of complex fractures in three dimensions (3D) is still a challenging task. Herein, we resolve this problem using a three-dimensional fabric determination technique based on Fourier transform. In our approach, the reconstructed 3D dataset is divided into several overlapping voxel patches. Using the Fourier transform, each voxel patch is represented by a gradient tensor to quantitatively characterise the spatial gradient field within the patch. The local inhomogeneities induced by the fracture structure take the major responsibility for the detected anisotropy characteristics and are quantified by the ordering parameters. The fracture orientation is directly calculated from the gradient tensor. The spatial gradient field can be visualised using an ellipsoidal re-scaling scheme, which presents a good agreement with the fracture macrostructure. The accuracy of the technique is visually validated by comparing the three-dimensional and two-dimensional projection results. The matching of the statistical results between 2D and 3D confirms the reliability of the proposed technique.