Medical Image Segmentation Based On Wolf Pack Algorithm

Image segmentation technology is the basis of image applications,and the quality of segmentation directly affects the subsequent steps of image processing.In recent years,with the rapid development of computer technology,the imaging methods of medical images are more abundant.And images provide more information,which provides great value for clinical medicine.Otsu threshold method is the most simple,direct and fast segmentation method in all image segmentation methods.It has the advantages of simple method and easy implementation.However,for complex images,the speed and quality of segmentation are limited,so the application of threshold segmentation is still rare.In order to improve the accuracy and efficiency of segmentation,this paper studies the medical image segmentation method based on wolf pack algorithm.The main contents of this paper are as follows:1.The speed and quality of threshold segmentation are limited.In order to solve the problem,a threshold segmentation method based on gamma fitting and modified wolf pack algorithm(GFMWPA)is proposed.The fitting function is used to locally fit the gray histogram of medical image,and the error value is transferred to the wolf pack algorithm for global optimization.At the same time,the threshold dimension of the fusion is objectively judged.In order to avoid missing the optimal solution due to the large step size and the influence of the small step size on the optimization speed,the adaptive optimization step size is adopted.The optimization direction is increased by iteration,the search range is expanded,and the segmentation accuracy is further improved.The experimental results show that the segmentation quality and speed are improved.2.Artificial bee colony algorithm is an optimization algorithm which imitates bee's foraging behavior.It uses the difference between individuals and food richness to generate new individuals randomly.It has been proved in practical application that it has fast convergence speed and robustness.In addition,in the experimental process,we find that the excellent historical experience of the search algorithm has a certain impact on the later algorithm implementation process.Therefore,this paper proposes another algorithm: a threshold segmentation method based on artificial bee colony and wolf pack algorithm.By introducing a unique update strategy to wolf pack algorithm and making full use of historical experience to improve the optimization degree and convergence rate.The improved algorithm is applied to Otsu multi-threshold segmentation.The experimental results show that the proposed algorithm can achieve fast and accurate segmentation.