PARALLEL GENETIC ALGORITHMS IN THE OPTIMIZATION OF MORPHOLOGICAL FILTERS - A GENERAL DESIGN TOOL

Mathematical morphology has produced an important class of nonlinear f ilters. Unfortunately, design methods existing for these types of filt er tend to be computationally intractable or require some expert knowl edge of mathematical morphology. Genetic algorithms (GAs) provide usef ul tools for optimization problems which are made difficult by substan tial complexity and uncertainty. Although genetic algorithms are easy to understand and simple to implement in comparison with deterministic design methods, they lend to require long computation limes. But the structure of a genetic algorithm lends itself well to parallel impleme ntation and, by parallelization of the GA, major improvements in compu tation time can be achieved. A method of morphological tilter design u sing GAs is described, together with an efficient parallelization impl ementation, which allows the use of massively parallel computers or in homogeneous clusters of workstations. (C) 1997 SPIE and IS&T.