An Optimal Image Thresholding Using Genetic Algorithm

Image segmentation plays an important and basic role in image processing and pattern recognition. Its purpose is to separate areas that do not superpose each other and to obtain the interested target. During the past few years many algorithms for image segmentation have been proposed. The popular technique is the threshold segmentation because of its simplicity and efficiency. Genetic algorithm is the immediate search method that is based on the theory of evolution which natural selection mechanism, parallel and statistics. In this paper, an optimal image thresholding using Genetic Algorithm is proposed. Compared with traditional threshold methods, the proposed method has advantages that it can implement quickly optimal threshold and have good capability and stabilization. The results show that using the proposed method can obtain satisfactory segmentation effect and save the computational time.