Segmentation of Noisy Binary Images Containing Circular and Elliptical Objects using Genetic Algorithms

A segmentation technique basically divides the spatial domain, on which the image is defined in ‘meaningful’ parts or regions. The current approaches involving Genetic Algorithms (GAs) segment the regular-shaped images. In the proposed method, this drawback is overcome by applying GA to images containing circular and elliptical objects. The GA generated the initial population set randomly where each individual is a possible solution for image segmentation. The reproduction step of GA uses morphological operations at random. Over several generations, population is evolved to get near optimal results. The experimental results are presented for noisy images containing circular and elliptical objects. The results of the proposed method are compared with the standard image segmentation techniques. The proposed method enhances the image segmentation for higher density noise level. General Terms Image Processing, Pattern Recognition.