Combination Clustering Algorithm Based On Density And Grid And Its Application In Image Segmentation

Image segmentation is an image processing technique which can divide the image into a number of significant areas according to the needs. Image segmentation also affect the quality of follow-up image processing.There are many methods for image segmentation,and some can be applied to any image,while others can be only used to special categories.In this paper,the density and grid clustering algorithm for image segmentation is applied to special categories for image.In this paper,density and grid clustering for image segmentation is main line,and relevant aspects are as follows:In the first chapter, In the first chapter, we do a comprehensive overview for the main cluster analysis methods existing in the literature. We analyze a variety of advantages and disadvantages of the algorithms for partitioning methods, hierarchical method, density-based method, grid-based methods and model-based method.The second chapter describes density and grid clustering algorithm..The chapter is divided into three sections to introduce density clustering algorithm,grid clustering method and density based on grid clustering algorithm.For density clustering algorithm,we mainly introduce traditional DBSCAN clustering method,fast FDBSCAN clustering algorithm,proposed IF-DBSCAN clustering method to improve previous two clustering algorithm,proposed DBSCAN clustering method based on data partition,and numerical experiments related to these density clustering algorithms which belong to the same strain.while for grid method,we mainly introduce three typical clustering algorithm including STING,WaveCluster and CLIQUE.In addition,we also present simulation associated with the clustering algorithms.Based on density clustering algorithm and grid clustering method,this chapter focuses on density clustering algorithm based on grid.For this clustering method,we mainly discuss DFC and GDCAP which are both typical density clustering algorithm based on grid.Chapter 3 present a sort of density clustering algorithm based on adaptive grid, taking into account the density clustering algorithm based on grid is not so perfect.Under the guidance of this theory,the author proposes a grid-based adaptive clustering algorithm for DBSCAN,and makes corresponding numerical experiments.Numerical experiments show that the proposed DBSCAN clustering method based on adaptive grid is feasible.The fourth chapter describes a image segmentation algorithm called DFC which is based on density and grid clustering.After this,analyze and compare DFC with FCM and brFCM in application on image segmentation.Comparative numerical experiments show that the performance of image segmentation algorithm called DFC which is based on density and grid clustering is good.Conclusion which is concise summarizes and discuss the main work in this paper.