Abstract
This paper proposes a method to extract plural segments in binary images. The genetic algorithm is applied that the chromosome is represented by the start address and the last address of a segment. The fitness of an individual is evaluated by the matching degree in a distance transformed image. However, the general genetic algorithm can extract only one segment in an image. To solve the problem, the algorithm of the immune system that can search plural local solutions is applied in the proposed method. The population is evolved according to the immunity of each individual. The experimental results show that the proper plural segments with free locations and lengths were extracted successfully by the proposed method because the method can search plural local solutions by keeping varieties of individuals.
