.1078, 北京科技大学学报 第30卷 像素灰度的空间信息(即像素灰度间的空间关系), [4]Gorriz J M,Ramlrez J.Lang E W,et al.Hard C-means cluster- 因而对噪声比较敏感:另一方面没有考虑不同样本 ing for voice activity detection.Speech Commun,2006.48: 1638 数据对聚类效果的不同影响,为了能够在算法中顾 [5]Chuang K S.Tzeng HL.Chen S W,et al.Fwzy C-means clus- 及像素灰度的空间信息的同时考虑不同样本数据对 tering with spatial information for image segmentation.Comput 聚类效果的不同贡献,本文在Szilagyi等人提出的 Med Imaging Graphics.2006.30:9 基于像素灰度空间信息的聚类算法EnFCM的基础 [6]Li J.Gao X B.Jiao LC.A new feature weighted fuzy clustering 上,通过引入基于灰度图像的直方图加权而修改了 algorithm.Acta Electron Sin.2006.34 (1):89 EFCM中的目标函数,从而提出了一种用于图像 (李洁,高新波,焦李成。基于特征加权的模糊聚类新算法 电子学报,2006,34(1):89) 分割的FCM聚类改进算法,实验结果表明,该改进 [7]Chen S C.Zhang D Q.Robust image segmentation using FCM 算法在分割含有不同类型噪声的图像时,均显示出 with spatial constraints based on new kernel-induced distance mea- 了该改进算法的优良性能 sure.IEEE Trans Syst Man Cybern Part B.2004.34 (4): 1907 参考文献 [8]Szilagyi L,Benyo Z.Szilagyii S M.et al.MR brain image seg" [1]Cheng H D.Jiang X H.Sun Y,et al.Color image segmentation mentation using an enhanced fuzy C-means algorithm//25th An- advances and prospects.Pattern Recognit.2001.34:2259 nual International Conference of IEEE EMBS.Piscataway, [2]Pham D L.Prince JL.An adaptive furzy C-means algorithm for 2003:17 image segmentation in the presence of intensity inhomogeneities. [9]Ahmed M N.Yamany S M,Mohamed N.A modified fuzzy Pattern Recognit Lett.1999.20:57 C means algorithm for bias field estimation and segmentation of [3]Chen W J.Giger M L.Bick U.A fuzzy C-means (FCM)-based MRI data.IEEE Trans Med Imaging.2002.21:193 approach for computerized segmentation of breast lesions in dy- [10]Zhang D Q.Chen S C.A novel kernelized furzy C-means algo- namic contrast-enhanced MR images.Acad Radiol,2006,13 rithm with application in medical image segmentation.Artif In- (1):63 tell Med,2004,32:37像素灰度的空间信息(即像素灰度间的空间关系) 因而对噪声比较敏感;另一方面没有考虑不同样本 数据对聚类效果的不同影响.为了能够在算法中顾 及像素灰度的空间信息的同时考虑不同样本数据对 聚类效果的不同贡献本文在 Szilagyi 等人提出的 基于像素灰度空间信息的聚类算法 EnFCM 的基础 上通过引入基于灰度图像的直方图加权而修改了 EnFCM 中的目标函数从而提出了一种用于图像 分割的 FCM 聚类改进算法.实验结果表明该改进 算法在分割含有不同类型噪声的图像时均显示出 了该改进算法的优良性能. 参 考 文 献 [1] Cheng H DJiang X HSun Yet al.Color image segmentation: advances and prospects.Pattern Recognit200134:2259 [2] Pham D LPrince J L.An adaptive fuzzy C-means algorithm for image segmentation in the presence of intensity inhomogeneities. Pattern Recognit Lett199920:57 [3] Chen W JGiger M LBick U.A fuzzy C-means (FCM)-based approach for computerized segmentation of breast lesions in dynamic contrast-enhanced MR images. Acad Radiol200613 (1):63 [4] Gorriz J MRamlrez JLang E Wet al.Hard C-means clustering for voice activity detection. Speech Commun200648: 1638 [5] Chuang K STzeng H LChen S Wet al.Fuzzy C-means clustering with spatial information for image segmentation.Comput Med Imaging Graphics200630:9 [6] Li JGao X BJiao L C.A new feature weighted fuzzy clustering algorithm.Acta Electron Sin200634(1):89 (李洁高新波焦李成.基于特征加权的模糊聚类新算法. 电子学报200634(1):89) [7] Chen S CZhang D Q.Robust image segmentation using FCM with spatial constraints based on new kerne-l induced distance measure.IEEE T rans Syst Man Cybern Part B200434 (4): 1907 [8] Szilagyi LBenyo ZSzilagyii S Met al.MR brain image segmentation using an enhanced fuzzy C-means algorithm∥25th A nnual International Conference of IEEE EMBS.Piscataway 2003:17 [9] Ahmed M NYamany S MMohamed N.A modified fuzzy C-means algorithm for bias field estimation and segmentation of MRI data.IEEE T rans Med Imaging200221:193 [10] Zhang D QChen S C.A novel kernelized fuzzy C-means algorithm with application in medical image segmentation.A rtif Intell Med200432:37 ·1078· 北 京 科 技 大 学 学 报 第30卷