Point Matching With a "Soft" Relaxation Algorithm

Abstract

Point-matching involves the matching of pairs of points from two sets of partially correlated points. It is an important task which is used in many different areas of signal processing. Although it is possible to perform point-matching using a brute-force algorithm, the high computational complexity makes it unfeasible even for a moderate number of points. In these circumstances an iterative relaxation algorithm is widely used. The traditional relaxation algorithm works well as long as the number of points in one set which do not have a corresponding pair in the second set is small and the positions of all the points are accurately known. When these conditions do not hold, the performance of the relaxation algorithm is substantially reduced. In this paper we formulate a "soft" relaxation algorithm using the concept of fuzzy linguistic quantifiers. The performance of the new relaxation algorithm is found to consistently exceed that of the traditional relaxation algorithm.