This paper describes a novel approach to relational matching problems
in machine vision. Rather than matching static scene descriptions, the
approach adopts an active representation of the delta to be matched,
This representation is based on a Delaunay triangulation that is itera
tively reconfigured to increase its degree of topological congruency w
ith the model relational structure in a reconstructive matching proces
s, The active reconfiguration of relational structures is controlled b
y a MAP update process. The final restored graph representation is opt
imal in the sense that it has maximum a posteriori probability with re
spect to the available attributes for the objects under match. The ben
efits of the technique are demonstrated experimentally on the matching
of cluttered synthetic aperture radar data to a model in the form of
a digital map. The operational limits of the method are established in
a simulation study. (C) 1998 Academic Press.