A set of X-ray tomographic images of a highly porous material composed of air-saturated coconut fibers isconsidered and used to estimate intrinsic characteristics of the material. Two physical properties are ofinterest: porosity and static airflow resistivity. For the porosity, two-dimensional...

A set of X-ray tomographic images of a highly porous material composed of air-saturated coconut fibers isconsidered and used to estimate intrinsic characteristics of the material. Two physical properties are ofinterest: porosity and static airflow resistivity. For the porosity, two-dimensional gray-scale tomographyimages are obtained and post-processed to produce approximative black and white ones, unambiguouslyattributing distinct regions to fibers and air locations. The porosity is then directly deduced counting theblack and white pixels. Several image processing algorithms are compared and associated porositiesrange from 0:76 to 0:97, depending on the method employed, while it is estimated to be 0:86 from ouranalysis. For the airflow resistivity, the idea followed here is to use the pattern of the post-processedimages as the lattice in a Lattice Boltzmann Method (LBM) fluid dynamics computation. To our knowl-edge, the LBM has not been used in this context before. A new 2D implementation of the method is there-fore developed and studied. After tuning computational parameters, we have estimated the airflowresistivity using ten images of our sample to be 1382#12 Pa:s=m2. Both porosity and resistivity resultsare fully consistent with measurements obtained from a porosity-meter and a resistivity-meter, demon-strating the pertinence of X-ray tomography and the associated proposed methods

Fechado