Image Segmentation Method Based on Particle Swarm Optimization and Spatial Distance Measure Clustering

Abstract

The invention discloses an image segmentation processing method based on particle swarm optimization and spatial distance measurement clustering, which mainly solves the problems of local misclassification and many regional noise points in the existing clustering image segmentation technology. The implementation steps are: (1) input the original image, extract the pixel features, and perform watershed segmentation; (2) calculate the adjacency matrix and generate clustering data according to the segmented area; (3) use the clustering data to randomly initialize the population; ( 4) Calculate the membership matrix and fitness value of the population, upgrade the individual optimal and global optimal, and evolve the population, (5) update the number of iterations, if the preset maximum number of iterations is reached, then output the best membership matrix , otherwise continue to step (4); (6) mark according to the principle of maximum probability according to the best membership degree matrix, and obtain the segmentation result. Compared with the prior art, the invention has good regional consistency and high segmentation accuracy, and can be used for target recognition of SAR images.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a segmentation method involving texture images and SAR images, which can be applied to target recognition. Background technique [0002] Image segmentation is one of the key technologies of image processing. The result of image segmentation is to divide the image into several parts, each part represents a different feature in the image, and mark the same part of pixels as the same value. The existing image segmentation methods mainly include methods based on regions, methods based on edge detection, methods based on clustering and so on. At present, people mostly use methods based on cluster analysis for image segmentation. The method based on cluster analysis is to use the known training sample set to find the decision-making classification points, lines or surfaces in the feature space of the image, and then map them back to the original image space to realize the di...

AI Technical Summary

Problems solved by technology

However, in the clustering process, because only the attributes of the region are considered, such as grayscale, texture, etc., the relationship between regions is ignored, and the integrity of spatial information is lacking, resulting in more noise points in the region and consistent regions. Poor performance and unsatisfactory segmentation

Images