Probabilistic finite modeling of stochastic estimation of image inter-frame geometric deformations

Abstract

The method of probabilistic finite modeling of stochastic estimation of image inter-frame geometric deformations parameters is proposed. The method is based on the discretization of the domain of deformation parameters and determination of estimates drift probability vector at each iteration (improvement, deterioration and no change of the vector with respect to coordinates of optimal values in the parameter space). The models of analyzed images and noises are given by the probability densities and the autocorrelation functions. It is a feature of the method. The usage of the probability distribution method for estimating the image deformations parameters generated by a non-identification relay procedure for a given finite number of iterations is considered. The mean square of error and the inter-frame correlation coefficient were used as cost estimation functions. Examples confirming the adequacy of the developed probabilistic mathematical model are presented.