Structural optimization of pseudogradient algorithms for estimating image parameters

The reliability of pattern-match searching for the fragment on image using set of pseudo-gradient procedures

“…If all search procedures are working in same conditions and will make equal number of iterations than it require huge computational expenses. To reduce these expenses the algorithm of managing ensemble of PGP [12] can be used. In this case on the each step priority of making next iteration is giving to procedure that has best value of some penalty function.…”

“…To reduce computational costs in [11] the algorithm for manipulating of set of PGP's is proposed where on the each step the priority to make next iteration to procedure that has best value of some penalty function (PF) X [12]. In this case step of the algorithm means complex of the operations: making ordinary iteration by procedure with best PF value, calculating new PF value and finding procedure with best PF value.…”

Pseudogradient Estimation of Digital Images Interframe Geometrical Deformations

“…To reduce computational expenses the following principle of structural optimization can be used (pseudogradient procedures set control). At each step of the algorithm the priority of the current iteration realization is given to the procedure, having the least value of a certain penalty function ψ characterizing the level of the priority (Tashlinskii, 2006). Here, by «algorithm step» we imply a set of operations that includes performing of the current iteration by the procedure with the least penalty function, finding …”