Probability graphic model image segmentation method based on flag fusion

Abstract

The invention relates to a probability graphic model image segmentation method based on flag fusion. The method comprises the steps of (1) acquiring a target image and multiple training images, conducting registration on the target image and the training images, and obtaining a corresponding deformation field; (2) acquiring the manual segmentation flag image of each training image subjected to registration, and obtaining a corresponding candidate flag according to deformation field reflection; (3) establishing a probability graphic model according to the training images and target image subjected to registration; (4) calculating the flag value of each pixel point in the target image by means of the local weighted voting flag fusion algorithm and the maximum posterior probability according to the probability graphic model, so that segmentation of the target image is achieved. Compared with the prior art, the method has the advantages that segmentation can be achieved automatically and segmentation precision is high.

Description

technical field [0001] The invention relates to the field of image processing, in particular to an image segmentation method of a probability graphic model based on marker fusion. Background technique [0002] The precise segmentation of specific tissues in human brain magnetic resonance (MR) images is a hot issue in the field of medical image processing, which is of great significance to the clinical diagnosis and research of related diseases. [0003] Brain tissue segmentation methods can be divided into manual, semi-automatic and fully automatic methods according to the degree of manual participation. The manual segmentation method can get very good expected results, but the disadvantage is low efficiency and poor repeatability. This method is usually only used to establish the gold standard of segmentation results; the semi-automatic segmentation method can reduce the workload to a certain extent and can combine experts. Prior knowledge can obtain segmentation results t...

AI Technical Summary

Problems solved by technology

The manual segmentation method can get very good expected results, but the disadvantage is low efficiency and poor repeatability. This method is usually only used to establish the gold standard of segmentation results; the semi-automatic segmentation method can reduce the workload to a certain extent and can combine experts. Prior knowledge can obtain segmentation results that are relatively close to the gold standard, but manual intervention (such as setting parameters) is required, and different parameter settings will lead to different segmentation results; the fully automatic segmentation method is independently completed by the computer, which has the advantages of no manual intervention, can Reproducing the characteristics of segmentation results has become the main research direction and development trend of segmentation methods

[0004] Most of the existing similarity measurement methods are based on the gray information of the image, without considering the regional characteristics of the image, and cannot well reflect the local similarity between the training image and the target image in the region where the segmented object is located.

When using this type of similarity as a reference image selection criterion, some training images that have a large similarity with the target image in the local area of ​​the segmentation object but have large differences in other areas may be missed, thus affecting the segmentation of the multi-training image method. precision

Images