基于菱形邻域和二维熵的改进密度峰值图像分割  

Improved Density Peak Image Segmentation Based on Two-dimensional Entropy of Diamond Neighborhood

作　　者：胡洋洋 刘增力[1] HU Yangyang;LIU Zengli(Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,China)

机构地区：[1]昆明理工大学信息工程与自动化学院,云南昆明650500

出　　处：《昆明理工大学学报（自然科学版）》2020年第2期66-73,共8页Journal of Kunming University of Science and Technology(Natural Science)

基　　金：国家自然科学基金项目(61271007).

摘　　要：自密度峰值聚类算法在2014年被提出以来,已经广泛应用在数据挖掘领域.针对聚类算法进行图像分割时,存在聚类数目稳定性问题和抗噪声性问题,本文在密度峰值聚类的基础上,引入菱形邻域信息和图像二维熵信息,充分利用图像的位置信息,进行单目标图像分割,使算法的抗噪性能增强和聚类数目更加稳定.实验结果验证了改进后的算法,在含有高斯噪声的图像和真实的图像的仿真上,具有较好的分割性能和抗噪性能.The density peak clustering algorithm has been widely used in data mining since proposed in 2014.In order to solve the problems of clustering number stability and resistance to noise of Clustering algorithm for image segmentation,this paper introduced the diamond neighborhood information and two-dimensional entropy image information based on the peak density clustering.Making full use of the position information of the image,we made single target image segmentation to boost the anti-noise performance of the algorithm and stabilize clustering number.Experimental results show that the improved algorithm has good segmentation performance and anti-noise performance in the simulation of images with Gaussian noise and real images.

关 键 词：密度峰值聚类 图像分割 菱形邻域 二维熵 

分 类 号：TP37[自动化与计算机技术—计算机系统结构]