Accurate Three‐dimensional Measurements of Features in Geological Materials from X‐ray Computed Tomography Data

The challenge of making accurate dimensional measurements of features in CT data volumes is best understood using the concept of the point-spread function (PSF). The PSF summarizes the blurring caused by the various non-idealities inherent in CT data acquisition, such as finite-sized source and detectors, motion during scanning, and reconstruction algorithms. The effect of these factors is to disperse the attenuation signal of any given feature into its immediate surroundings in a CT data slice. By identifYing and quantifying the attenuation signal of a feature, precise measurements can be made, even on features with one or more dimensions smaller than a data voxel. This type of calculation has been used to successfully measure fracture widths, and can be applied to volumetric measurements as well. The common method of delineating objects using thresholds based on the FWHM (full width half max) criterion is appropriate for features larger than the PSF diameter, but leads to erroneous results if one or more feature dimensions are smaller.