Medical image segmentation method based on improved range adjustment level set algorithm

Abstract

The invention discloses a medical image segmentation method based on an improved range adjustment level set algorithm. The method comprises the following steps that a two-dimensional medical image is input; Gaussian filter smoothing is carried out on the input two-dimensional medical image in a two-dimensional space through Matlab software and a Gaussian kernel function, the gradient information of the image processed through smooth filtering is , the contour curve of the two-dimensional medical image is obtained, an ADRLSE model is obtained by improving an implicit function of a DRLSE model, and an edge stopping function of Neumann conditions is set through the ADRLSE model; the contour curve can adjust the evolution direction in a self-adaptation mode through evolution of the contour curve till stopping conditions are met. According to the medical image segmentation method based on the improved range adjustment level set algorithm, the attributes of a signed distance function are kept, re-initialization is avoided, the calculated amount is reduced, the problem of periodic initialization of a sign function is solved, the efficiency of the algorithm is improved, and edge distortion of image segmentation is avoided.

Description

technical field [0001] The invention relates to a medical image segmentation method. In particular, it concerns an improved distance-adjusted level set method. Background technique [0002] Image segmentation is one of the classic research topics in the fields of image processing, image analysis and computer vision, and it is also one of the difficulties. Provide basic support for quantitative and qualitative analysis. [0003] Due to the different methods of various images, especially medical images, certain random noise is inevitably introduced in the image processing process, and the shapes and intensities of the objects of interest in the images are very different, so far there is no general The method is suitable for all image segmentation. The level set method developed in recent years comprehensively utilizes region and boundary information, and is widely used in image segmentation and computer vision because of its accuracy, automation and continuity of final segm...

AI Technical Summary

Problems solved by technology

Li et al. introduced a signed distance function to punish the energy functional, that is, a level set method without initialization, which solves the problem of periodic initialization of the sign function and improves the efficiency of the algorithm, but boundaries often appear when segmenting images with blurred or discontinuous boundaries. Leakage phenomenon

Images