Image segmentation method based on immune clone algorithm and fuzzy kernel-clustering algorithm

Abstract

The invention discloses an image segmentation method based on an immune clone algorithm and a fuzzy kernel-clustering algorithm. The image segmentation method aims to solve the technical fields of sensitive initial clustering center, easy local optimum falling, and low segmentation correct rate of a fuzzy C-means clustering method. The image segmentation method includes the steps of reading an image to test the image size and convert the image to a grey-scale image; setting the parameters of a fuzzy kernel-clustering method; filtering the grey-scale image by a non-local means filtering method capable of automatically adjusting parameters to obtain the filtered image; optimizing the grey-scale image of the input image by means of the immune clone algorithm to obtain an initial clustering center; obtaining a final membership grade matrix uki by means of membership grade matrix formula with filtering and obtaining the final clustering center value v2 by means of a clustering center formula with filtering; and outputting the segmented image by means of fuzzification. The image segmentation method has the advantages of local optimum falling free, high segmentation accuracy, and good robustness, and can be applied to the segmentation of artificial synthesis images, medical images natural images or other images.

Description

technical field [0001] The invention belongs to the technical field of image processing, and mainly relates to image segmentation, in particular to an image segmentation method based on immune cloning and fuzzy kernel clustering, which can be used for segmentation of various images such as artificially synthesized images, medical images, and natural images. Background technique [0002] Image segmentation is one of the important contents in the field of computer vision, the primary problem of pattern recognition, and a classic problem of image processing. Objects can be extracted and recognized only on the basis of image segmentation, and the quality of image segmentation directly affects higher-level image analysis and understanding. Therefore, the research on image segmentation is of great significance. [0003] At present, many image segmentation methods based on fuzzy C-means clustering (FCM) have been proposed. FCM is suitable for the characteristics of uncertainty and...

AI Technical Summary

Problems solved by technology

[0003] At present, many image segmentation methods based on fuzzy C-means clustering (FCM) have been proposed. FCM is suitable for the characteristics of uncertainty and fuzziness in images. It is an unsupervised classification method, and the segmentation process does not require any manual intervention. , suitable for the field of automatic segmentation, so the FCM segmentation method has the characteristics of strong adaptability and fast convergence speed. The disadvantage of FCM is that the FCM algorithm is sensitive to the initial clustering center value, and it is easy to converge to the local optimal value. Does not consider the spatial domain information of pixels, is sensitive to noise, and has low robustness

Compared with the traditional fuzzy C-means clustering algorithm, this method introduces a weighted fuzzy factor into the objective function, which contains the domain information of the image, so that more details of the image are considered in the segmentation process. The initial clustering center used is given by random initialization. The initial clustering center obtained by random initialization makes the update process of the clustering center trapped in a local optimum, and the global optimal clustering center cannot be obtained, and this method only considers The neighborhood information of the image is not considered, and the non-local information of the pixels of the image is not considered, so the segmentation rate of the noisy image is low and the robustness is poor.

The distance measure in the membership degree formula of this method is Euclidean distance, which is a distance measure with poor robustness, so it is still sensitive to noise points, poor robustness, low segmentation accuracy, and the initial clustering The center is also randomly given, which makes it easy to fall into local optimum during the update process of the cluster center

Images