Grain Partitioning and its Applications

In this paper we introduce a new robust technique for the analysis of rock core fragments and granular materials from 3D digital images. We used a high resolution X-ray tomography apparatus to image the materials. We segment the density attenuation map into distinct pore and solid (grain) phases by applying a number of image processing and analysis steps [SHE 04]. Using the segmented image, we identify individual grains using a fully paralleled partitioning algorithm, which is based on the assumption that the boundaries between touching grains coincide with watershed surf aces of the Euclidean distance function. This technique is a powerful tool for particle characterisation of granulated materials. It allows a broad range of studies from fundamental studies of granular packings to applied studies of subsidence in oil reservoirs.