Estimating the Parameters of a Stochastic Geometrical Model for Multiphase Flow Images Using Local Measures

This paper presents a new method for estimating the parameters of stochastic geometric model multiphase flow image processing using local measures. Local measures differ from global in that they are only based on small part binary and consequently provide different information certain properties such as area perimeter. Since have been shown to be helpful typical grain elongation ratio homogeneous Boolean model, objective this study was use these statistically infer more complex non-Boolean sample observations. An optimization algorithm is used minimize cost function likelihood probability densityof measurements. The performance analysed numerical experiments real errors relative images most model-generated less than 2%. covariance particle size distribution also calculated compared.