A K-nearest neighbor and multi-class merge based density peak clustering method and image segmentation system

Abstract

A K-nearest neighbor and multi-class merge based density peak clustering method and image segmentation system are disclosed, which use density calculation mode to describe sample distribution and adopt new evaluation index to obtain clustering center. Iterative allocation strategy is designed to classify the remaining points accurately. A local class merge method is given to prevent class splitting that will contain multiple density peak points. The invention describes the distribution of each data point through the density measurement method, constructs a preference index based on the densityand the distance to evaluate the clustering center, and utilizes the iterative distribution strategy to allocate the remaining points to improve the clustering accuracy. After completing local clustering, multi-class merging strategy is used to complete the local class merging operation. The results of numerical experiments show that it has good applicability in 22 test data sets and real data.

Description

technical field [0001] The invention belongs to the technical field of reading or recognizing printed or written characters or used to recognize graphics, and in particular relates to a clustering method based on K-nearest neighbors and multi-category merged density peaks, and an image segmentation system. Background technique [0002] At present, the existing technologies commonly used in the industry are as follows: clustering can mine potentially valuable information from unordered data, and it has broad application prospects in many fields such as image segmentation, document restoration, and pattern classification. At present, there are many clustering methods, including segmentation clustering, hierarchical clustering, density clustering and grid-based clustering. K-means is the simplest and most popular segmentation clustering algorithm, which has the advantages of simple operation and fast speed, but it is very dependent on the number of clusters and the initial clus...

AI Technical Summary

Problems solved by technology

[0003] To sum up, the problem existing in the existing technology is: when the density peak clustering algorithm is dealing with data sets with complex structure, high dimensionality and multiple density peaks in the same category, due to the local density measurement method adopted by the method is not Uniform, and it is easy to cause error propagation when assigning the remaining points and clustering the clusters containing multi-density peaks into multiple clusters, so that the clustering quality obtained by this algorithm is low, and it is difficult to effectively apply to practical problems

Images