空间加权模糊C均值聚类图像分割算法  

Spatial Weighted Fuzzy C-means Clustering Algorithm for Image Segmentation

作　　者：李小和[1] 屈展[1] 王魁生[1] 卢胜男[1] LI Xiaohe QU Zhan WANG Kuisheng LU Shengnan(College of Computer Science, Xihn Shiyou University, Xi＇an 710065, Shaanxi, China)

机构地区：[1]西安石油大学计算机学院,陕西西安710065

出　　处：《西安石油大学学报（自然科学版）》2017年第5期102-106,共5页Journal of Xi’an Shiyou University（Natural Science Edition）

基　　金：国家自然科学基金项目(51674200);西安石油大学青年科技创新基金项目(2013BS021)

摘　　要：为了提高模糊C均值聚类(FCM)算法用于图像分割时对噪声的鲁棒性,在FCM算法中引入了图像像素的邻域约束,提出一种空间加权模糊C均值聚类图像分割算法。首先根据邻域像素的模糊隶属度函数值,定义像素分类标记的局部先验概率,然后将该局部先验概率融入标准的FCM算法的目标函数中,从而提出一种空间加权模糊C均值聚类图像分割算法。仿真实验通过合成图像和真实图像验证了该算法的有效性和鲁棒性。Fuzzy C-means clustering（ FCM） is an unsupervised clustering method,which is widely used in image segmentation. A spatial weighted fuzzy C-means clustering algorithm for image segmentation is proposed in order to overcome the sensitivity of the standard FCM algorithm to noises and other imaging artifacts. Firstly,the local prior probabilities of pixel classification marks are defined are defined according to the fuzzy membership function value of neighborhood pixels,and then those local prior probabilities are incorporated into the objective function of the standard FCM. Simulation experiments show the effectiveness and robustness of the proposed algorithm through both synthetic and real images.

关 键 词：图像分割算法 模糊C均值聚类 空间加权 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]