Genetic algorithm approach to image segmentation using morphological operations

This paper presents an approach for image segmentation using genetic algorithms (GA) in conjunction with morphological operations. The GA starts with a population of solutions, initialized randomly, to represent possible segmentations of the image. The solutions are evaluated using an appropriate fitness function and the fittest candidates are selected to be parents for producing offsprings that form the next generation. Morphological operations are applied in the reproduction step of the GA to exploit a priori image information. Over several generations, populations evolve to yield the optimal results. The feasibility of applying genetic algorithms to image segmentation is investigated and initial results of segmentation of noisy images are presented.