邻域灰度差加权的模糊C均值聚类图像分割算法  

An Image Segmentation Algorithm of Weighted with Neighborhood Gray Difference Fuzzy C-means Clustering

作　　者：沙秀艳[1,2] 何友[2] 王贞俭[1] 

机构地区：[1]鲁东大学数学与信息学院,山东烟台264025 [2]海军航空工程学院,山东烟台264001

出　　处：《火力与指挥控制》2008年第12期34-36,40,共4页Fire Control & Command Control

基　　金：国家自然科学基金资助项目(60172033);鲁东大学校基金资助项目(L20072703)

摘　　要：模糊C均值(FCM)算法用于灰度图像分割是一种非监督模糊聚类后再标定的过程,适合灰度图像中存在着模糊和不确定性的特点。但是这种算法没有考虑到样本空间中不同的样本点对分类的贡献不同,因此分割效果不理想。提出了邻域灰度差加权的模糊C均值聚类算法,实验结果表明,该算法不仅取得了很好的分割效果,而且加快了算法的收敛速度,从而满足了图像分割的有效性、实时性的要求。It is a procedure of the label following an unsupervised fuzzy clustering that fuzzy c-means （FCM） algorithm is applied for gray image segmentation,and it suits for the uncertain and ambiguous characters in gray image. But this algorithm doesn＇t consider that different sample variously influences the result of the class in sample space. So the result isn＇t ideal. This paper introduces an image segmentation algorithm of weighted with neighborhood gray difference fuzzy C-means clustering. Experimental results demonstrate that this scheme can not only effectively segment, but also quikly converge. So it can effectively and timely segment the target from its background.

关 键 词：图象分割 模糊C均值聚类 加权模糊C均值聚类 邻域灰度差 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]