摘要
模糊C均值(FCM)聚类算法广泛用于图像的自动分割,但是该算法没有考虑像素的灰度和空间特征,对噪声十分敏感。因此提出一种改进的算法,在传统的FCM聚类的基础上,运用邻域像素的灰度相似度和聚类分布统计来构造新的隶属函数,对图像进行聚类分割。该方法不仅有效地抑制了噪声的干扰,而且把错分类的像素很容易的纠正过来。对两种类型的含噪图像的实验结果表明该方法对噪声具有很强的鲁棒性和对像素聚类的正确性。
Fuzzy c-means （FCM） clustering algorithm has been widely used in automated image segmentation, However, the conventional FCM algorithm is sensitive to noise because of taking no into account the gray and spatial information. An improved algorithm based on the preliminary image segmentation with the FCM cluster is proposed. The degree of gray similarity and cluster distribution statistics of the neighbor pixels are used to form a new membership function, It is not only effective to constrain the noise, but also ease to correct the misclassified pixels. Experimental results on two types of noisy images indicate that the segmentations are more accurate and robust than the standard FCM algorithm.
出处
《计算机工程与设计》
CSCD
北大核心
2007年第6期1358-1360,1363,共4页
Computer Engineering and Design
基金
甘肃省自然科学基金项目(3ZS042-B25-007)
