A brain MRI image segmentation method based on fuzzy multi-threshold and regional information

Abstract

Aiming at the influence of grayscale inhomogeneity and medical artifacts on the segmentation of the brain MRI images in the prior art, the present invention takes the combination of regional information aggregation and threshold segmentation as the design idea, introduces the fuzzy multi-threshold technology, and proposes a fuzzy multi-threshold based method. Brain MRI image segmentation method with regional information. On the basis of fuzzy multi-threshold segmentation, the method further improves the quality of image segmentation by constructing fuzzy membership function and fuzzy membership aggregation based on local information. The present invention adopts fuzzy theory and information aggregation technology to suppress uneven grayscale. The influence of sexual and medical artifacts on segmentation results can retain more original image information, effectively avoid false segmentation caused by artifacts, and improve the effect of brain MRI image segmentation. The present invention adopts an improved quantum particle swarm optimization algorithm, introduces an exponentially decreasing shrinkage-expansion coefficient, and enhances the search performance of the algorithm. At the same time, the convergence speed of the algorithm is improved.

Description

technical field [0001] The invention belongs to the technical field of image segmentation, relates to a fuzzy multi-threshold segmentation technology, and mainly relates to a brain MRI image segmentation method based on fuzzy multi-threshold and regional information. Background technique [0002] Image segmentation is the technique and process of dividing an image into several specific regions with unique properties and proposing objects of interest. It is a key step from image processing to image analysis. Compared with natural image segmentation, medical image segmentation often depends on specific applications, imaging modalities and specific body parts, which is a more complex problem. Many algorithms that perform well in natural image segmentation have problems such as under-segmentation and over-segmentation in medical image segmentation, so they cannot be transplanted well. For example, in the brain magnetic resonance imaging (Magnetic Resonance Imaging, MRI) image,...

AI Technical Summary

Problems solved by technology

Its disadvantage is that it only considers the influence of a single pixel on the segmentation result, and does not use the regional characteristics of the image, and the segmentation result is prone to isolated pixels and regions.

In addition, the traditional threshold segmentation method is not suitable for the gray level inhomogeneity and medical artifacts of brain MRI images.

Images