| Medical image segmentation is a hot topic in the field of image segmentation.Medical image segmentation is to segment medical image into multiple regions and extract useful targets.Many methods have been developed over the years.Because of the difference of blood vessel tissue of human body,the medical image segmentation requires high accuracy and accuracy,and because the image is easy to be affected by external factors such as noise,it is difficult for the general segmentation algorithm to meet the requirements.Tomography is a kind of unsupervised clustering model,which is very classical in the field of image segmentation,and has been paid more and more attention.The work carried out in this paper is as follows:1.Based on the literature of image segmentation,the tomography and EM algorithms are analyzed.The innovation of this paper lies in the application of Gao Si distribution,tomographic imaging algorithm and EM algorithm.In this paper,the EM algorithm and CT algorithm are combined with computed tomography imaging algorithm,point algorithm,image restoration algorithm and so on.Point algorithm,gray processing algorithm,threshold processing algorithm,linear transformation algorithm,pseudo-color processing algorithm and other synthetic algorithm practice is also the innovation of this paper.2.In this paper,a segmentation method based on tomography is designed to segment the coronary artery in CT image.EM algorithm is a common algorithm for estimating the parameters of CT scan imaging,so it is introduced to the estimation of parameters in CT imaging.Finally,the segmented image is processed by the connected domain labeling algorithm.In this method,the pixel points in the image are grouped into two categories by tomography,and the iteration times are tested from 200 times to 150 times,100 times to 50 times and 50 times to 10 times.About 300 images are segmented.Some typical image segmentation results are selected from the images to be displayed,and the segmented images are displayed.The image is evaluated.Finally,the results show that the proposed algorithm can effectively separate coronary artery from other tissues in CT images.3.In this paper,300 vascular images are segmented,and the segmentation effect is very good.Finally,the cardiovascular image in the image can be obtained by post-processing the segmented image by using the connected domain algorithm.In the future research,the estimation of initial value can be calculated by better method,and the segmentation result can be further enhanced.In this paper,the imaging functions such as industry,geophysics,engineering,agriculture,safety detection and so on have been widely used by using certain mathematical methods and computer processing,which has certain theoretical and practical significance. |