Self-adaptive spinal CT image segmentation method based on particle swarm optimization

Abstract

The invention relates to a self-adaptive spinal CT image segmentation method based on particle swarm optimization. The method comprises the following steps that (1) an original image is inputted; (2) the first generation of population is initialized; (3) an individual optimum and a global optimum are calculated; (4) new individuals are generated; (5) a new individual optimum and a global optimum are calculated; (6) whether the maximum number of iterations is met is judged, and the process goes to step (7) if the judgment result is yes, or the process returns to the step (4); (7) initial image segmentation is performed to act as an initial segmentation result; and (8) topology operation is adopted to perform further accurate spinal segmentation so that the image is outputted. Search granularity is optimized, and speed of algorithm convergence is controlled in a coarse-to-fine way so that speed of algorithm convergence is greatly enhanced, segmentation precision is enhanced, further accurate spinal segmentation is performed by adopting the topology through the prior knowledge of areas to be segmented, and thus image segmentation efficiency and quality are greatly enhanced.

Description

technical field [0001] The present invention relates to general image processing, specifically to image analysis, and in particular to image segmentation through analysis of image attributes. Background technique [0002] With the continuous development of computer science and technology, image processing technology has also been developed to a certain extent, and gradually formed its own scientific theory system. Image segmentation is an important technology of image processing, which has been paid more and more attention by more and more people in practical application and theoretical research. [0003] Image segmentation technology is to divide the image into several meaningful regions and extract the objects of interest so that the follow-up work can be carried out effectively. It is a key step from image processing to image analysis. The existing image segmentation methods can be divided into the following categories: segmentation methods based on regions, segmentation...

AI Technical Summary

Problems solved by technology

However, although the particle swarm algorithm is simple to implement and has few parameters, it is easy to fall into local optimum, and the selection of parameters directly affects the speed and accuracy of image segmentation.

[0004] In "ImageThresholdSegmentationMethodBasedonImprovedParticleSwarmOptimization" (ComputerScience39(2012) 289-301), Zhang Hui and others proposed an image threshold segmentation method based on chaotic particle swarm optimization. The global search ability of the algorithm is improved, and it can avoid falling into the local optimal state to a certain extent, but the segmentation accuracy and time complexity of the algorithm are improved.

Images