A new robust algorithm analyzing the geometry and connectivity of the pore space of sedimentary rock is based on fundamental concepts of mathematical morphology. The algorithm distinguishes between the “pore bodies” and “pore throats,” and establishes their respective volumes and connectivity. The proposed algorithm also produces a stick-and-ball diagram of the rock pore space. The tests on a pack of equal spheres, for which the results are verifiable, confirm its stability. The impact of image resolution on the algorithm output is investigated on the images of computer-generated pore space.

One of distinctive features of our approach is that no image thinning is applied. Instead, the information about the skeleton is stored through the maximal inscribed balls or spheres (MIS) associated with each voxel. These maximal balls retain information about the entire pore space. Comparison with the results obtained by a thinning procedure preserving some topological properties of the pore space shows that our method produces more realistic estimates of the number and shapes of pore bodies and pore throats, and the pore coordination numbers.

The distribution of maximal inscribed spheres makes possible simulation of mercury injection and computation of the corresponding dimensionless capillary pressure curve. It turns out that the calculated capillary pressure curve is a robust descriptor of the pore space geometry and, in particular, can be used to determine the quality of computer-based rock reconstruction.