DISCRIMINATIVE GRAPHICAL MODEL FOR POROUS MEDIA IMAGE SYNTHESIS

Abstract

I maging synthesis methods open a new door to help scientists for furtherstudy on porous materials. H igh resolution images are required t o analyze the macroscopic properties of porous media. However, a few degenerated high resolution samples are available because ofconstraints, and low resolution measurements (such as MRI images) cannot fully describe themedium. Computer-aided approaches can help the science of porous media by generating manyartificial high resolution samples using the information of available data. In this paper, a noveldiscriminative graphical framework is proposed which statistically models the synthesis problem.The probability distribution of high resolution image of a porous medium given a low resolutionmeasurement is modeled by conditional random fields (CRF). A Monte Carlo approach isproposed to sample the constructed model and to generate high resolution samples. Moreover, ahierarchical CRF is proposed for gradual synthesis of high resolution porous media images. Thesuccess of the models is shown and compared through several experimental results.