An approach enabling to reduce computational expenses at pseudogradient estimation of image parameters based on control of goal function local sample volume is proposed. Local sample volume variation during the process of parameter estimation occurs automatically in correspondence with a preassigned criterion, for instance, sample correlation coefficient. It is shown that for the problems of image mutual spatial deformation parameter estimation, computational expenses can be reduced several times as many.

No abstract

No abstract

No abstract

No abstract

No abstract

No abstract

The approach to performance enhancement of pseudogradient search procedures by samples of fragments in images based on the analysis of convergence behavior of estimated locations of the sought frag ments is considered for the image matching method based on a set of estimated locations of predetermined local fragments. Stopping criteria are suggested for the location estimation of control fragments in the image matching problem based on the developed approach.