CT image segmentation method based on novel threshold value formula and self-adaptive double-potential-well function

Abstract

The invention relates to a CT image segmentation method based on a novel threshold value formula and a self-adaptive double-potential-well function, and belongs to the field of image processing. The method is divided into a coarse segmentation part and a fine segmentation part, a region growing model is taken as the coarse segmentation part, a threshold value formula based on average absolute deviation is provided, and the formula can automatically generate a reasonable threshold value according to image features and can process images with uneven gray levels or low contrast; an LRCV model is used as fine segmentation, a self-adaptive double-potential-well function is provided, coefficients are dynamically adjusted, so that the diffusion rate at the initial stage of segmentation is increased, the diffusion rate at the later stage of segmentation is reduced, the diffusion rate near a zero potential well is reduced, the possibility that a zero level set invades the interior of a segmented target is reduced, and the segmentation precision is improved. The problem that the initial contour of the LRCV model is sensitive is effectively solved. Compared with a traditional active contour model, the method has a more accurate segmentation result for a CT image with non-uniform gray scale or low contrast.

Description

technical field [0001] The invention belongs to the field of image processing and relates to a CT image segmentation method based on a novel threshold formula and an adaptive double potential well function. Background technique [0002] Computed Tomography (CT) obtains projection data by scanning an object with rays, and uses it to reconstruct a tomographic image of the object. Image segmentation is to divide the image into several disjoint regions with different characteristics, and extract and separate the target region of interest according to certain characteristics of the image on these regions. Image segmentation is a key step from image processing to image analysis and understanding. Whether the target of interest can be accurately and efficiently extracted from CT images is crucial to subsequent processing (such as defect detection, dimension measurement, and reverse manufacturing, etc.). Therefore, it is of great practical significance to study how to improve the a...

AI Technical Summary

Problems solved by technology

Combining region growth and active contours, the invention is divided into two parts: rough segmentation and fine segmentation. The region growth model is used as the coarse segmentation, and the LRCV model is used as the fine segmentation, which effectively solves the problem that LRCV is sensitive to the initial contour, and then improves Segmentation Accuracy of LRCV Model for CT Image

Images