An Improved Algorithm for Medical Image Segmentation

The main objective of medical image segmentation is to extract and characterize anatomical structures with respect to some input features or expert knowledge. Traditional two-dimensional Otsu method for medical image segmentation is time-consuming computation and become an obstacle in real time application systems. This paper describes a way of medical image segmentation using optimized two-dimensional Otsu method based on improved genetic algorithm (GA). In proposed algorithm, the probability-ties of crossover are adaptively varied depending on the ranking value of individuals instead of fitness, and dyadic mutation operator was presented to take the place of the traditional one. The experimental results show that the new optimized method dramatically reduces the operating time in medical image segmentation while ensures the final image segmentation quality.