Spatial clustering procedures for region analysis

摘要

This paper describes two clustering procedures for region analysis of image data and discusses the security of these algorithms theoretically. First our algorithms find kernels of regions and then classify pixels into regions using these kernels. The first algorithm distinguishes the regions that have far more distances than the given distance and the second algorithm distinguishes C regions that are great distances from each other in the feature space. These parameters are criteria which decide whether regions are similar or dissimilar. Examples are presented in order to show how these algorithms work for real image data.