第4期 朱占龙，等：融入类贡献抑制因子的灰度级模糊C均值图像分割 ·647· (a)#NDT5噪声图 (b)标准分割图 (c)FFCMAF算法 (d)NDFCM算法 (e)EnFCM算法 (f)FGFCM算法 (g)FNDFCM P算法 (h)FCMRFC算法 图7椒盐噪声SPN(10%)#粉NDT5图像的分割结果 Fig.7 Segmentation results of #NDT5 when SPN(10%) 4结束语 (natural science edition),2017,29(3):377-381 [5]王小鹏，张永芳，王伟，等.基于自适应滤波的快速广义 为了改善基于灰度级模糊C均值算法不能有 模糊C均值聚类图像分割[).模式识别与人工智能， 效分割类间差异较大图像的缺点，在该算法基础 2018,31(11):1040-1046 上提出一种改进版本。改进的算法主要是通过引 WANG Xiaopeng,ZHANG Yongfang,WANG Wei,et al. 入一种新颖的类贡献抑制因子至目标函数中，减 Image segmentation using fast generalized fuzzy C-means 弱较大类对目标函数的作用，避免较小类的聚类 clustering based on adaptive filtering[J].Pattern recogni- 中心向较大类偏移。基于新的目标函数对图像进 tion and artificial intelligence,2018,31(11):1040-1046. 行迭代聚类分割，结果显示新算法的有效性和鲁 [6]ZHAO Feng,FAN Jiulun,LIU Hanqiang.Optimal-selec- 棒性。需指出，改进算法如果在图像去噪阶段能 tion-based suppressed fuzzy c-means clustering algorithm 够得到更好的待分割图像，那么算法的指标值和 with self-tuning non local spatial information for image 视觉效果会进一步改善，故研究图像去噪算法仍 segmentation[J].Expert systems with applications,2014, 41(9):4083-4093. 具有较大意义。 [7]GUO Fangfang,WANG Xiuxiu,SHEN Jie.Adaptive 参考文献： fuzzy c-means algorithm based on local noise detecting for image segmentation[J].IET image processing,2016,10(4): [1]李玉，胡海峰，赵雪梅，等.遥感图像扫描聚类分割算 272-279. 法U.信号处理，2018,34(9)：1130-1141. [8]雷涛，张肖，加小红，等.基于模糊聚类的图像分割研究 LI Yu,HU Haifeng,ZHAO Xuemei,et al.The remote 进展U.电子学报，2019,47(8)：1776-1791 sensing image scan clustering segmentation algorithm[J]. LEI Tao,ZHANG Xiao,JIA Xiaohong,et al.Research Journal of signal processing,2018,34(9):1130-1141. Progress on Image Segmentation Based on Fuzzy Cluster- [2]WU Chengmao,YANG Xiaoqiang.Robust credibilistic ing[J].Acta electronica sinica,2019,47(8):1776-1791. fuzzy local information clustering with spatial information [9]FAN Jiulun,ZHEN Wenzhi,XIE Weixin.Suppressed constraints[J].Digital signal processing,2020,97:102615. fuzzy c-means clustering algorithm[J].Pattern recognition [3]CHAIRA T.A novel intuitionistic fuzzy C means cluster- letters.2003.24(9/10):1607-1612. ing algorithm and its application to medical images[J].Ap- [10]HUNG W L,YANG M S,CHEN Dehua.Parameter se- plied soft computing,2011,11(2):1711-1717. lection for suppressed fuzzy c-means with an application [4]丁晓峰，何凯霖.一种基于一致性分片FCM的图像分割 to MRI segmentation[J].Pattern recognition letters,2006. 算法.重庆邮电大学学报（自然科学版），2017,29(3)： 27(5):424-438 377-381 [11]SZILAGYI L,BENYO Z,SZILAGYI S M,et al.MR DING Xiaofeng,HE Kailin.A homogeneous pieces based brain image segmentation using an enhanced fuzzy c- FCM algorithm for image segmentation[J].Journal of means algorithm[C]//Proceedings of the 25th Annual In- Chongqing University of Posts and Telecommunications ternational Conference of the IEEE Engineering in Medi-(a) #NDT5 噪声图 (b) 标准分割图 (c) FFCMAF 算法 (d) NDFCM 算法 (e) EnFCM 算法 (f) FGFCM 算法 (g) FNDFCM_P 算法 (h) FCMRFC 算法 图 7 椒盐噪声 SPN(10%)#NDT5 图像的分割结果 Fig. 7 Segmentation results of #NDT5 when SPN (10%) 4 结束语 为了改善基于灰度级模糊 C 均值算法不能有 效分割类间差异较大图像的缺点，在该算法基础 上提出一种改进版本。改进的算法主要是通过引 入一种新颖的类贡献抑制因子至目标函数中，减 弱较大类对目标函数的作用，避免较小类的聚类 中心向较大类偏移。基于新的目标函数对图像进 行迭代聚类分割，结果显示新算法的有效性和鲁 棒性。需指出，改进算法如果在图像去噪阶段能 够得到更好的待分割图像，那么算法的指标值和 视觉效果会进一步改善，故研究图像去噪算法仍 具有较大意义。 参考文献： 李玉, 胡海峰, 赵雪梅, 等. 遥感图像扫描聚类分割算 法 [J]. 信号处理, 2018, 34(9): 1130–1141. LI Yu, HU Haifeng, ZHAO Xuemei, et al. The remote sensing image scan clustering segmentation algorithm[J]. Journal of signal processing, 2018, 34(9): 1130–1141. [1] WU Chengmao, YANG Xiaoqiang. Robust credibilistic fuzzy local information clustering with spatial information constraints[J]. Digital signal processing, 2020, 97: 102615. [2] CHAIRA T. A novel intuitionistic fuzzy C means cluster￾ing algorithm and its application to medical images[J]. Ap￾plied soft computing, 2011, 11(2): 1711–1717. [3] 丁晓峰, 何凯霖. 一种基于一致性分片 FCM 的图像分割 算法 [J]. 重庆邮电大学学报(自然科学版), 2017, 29(3): 377–381. DING Xiaofeng, HE Kailin. A homogeneous pieces based FCM algorithm for image segmentation[J]. Journal of Chongqing University of Posts and Telecommunications [4] (natural science edition), 2017, 29(3): 377–381. 王小鹏, 张永芳, 王伟, 等. 基于自适应滤波的快速广义 模糊 C 均值聚类图像分割 [J]. 模式识别与人工智能, 2018, 31(11): 1040–1046. WANG Xiaopeng, ZHANG Yongfang, WANG Wei, et al. Image segmentation using fast generalized fuzzy C-means clustering based on adaptive filtering[J]. Pattern recogni￾tion and artificial intelligence, 2018, 31(11): 1040–1046. [5] ZHAO Feng, FAN Jiulun, LIU Hanqiang. Optimal-selec￾tion-based suppressed fuzzy c-means clustering algorithm with self-tuning non local spatial information for image segmentation[J]. Expert systems with applications, 2014, 41(9): 4083–4093. [6] GUO Fangfang, WANG Xiuxiu, SHEN Jie. Adaptive fuzzy c-means algorithm based on local noise detecting for image segmentation[J]. IET image processing, 2016, 10(4): 272–279. [7] 雷涛, 张肖, 加小红, 等. 基于模糊聚类的图像分割研究 进展 [J]. 电子学报, 2019, 47(8): 1776–1791. LEI Tao, ZHANG Xiao, JIA Xiaohong, et al. Research Progress on Image Segmentation Based on Fuzzy Cluster￾ing[J]. Acta electronica sinica, 2019, 47(8): 1776–1791. [8] FAN Jiulun, ZHEN Wenzhi, XIE Weixin. Suppressed fuzzy c-means clustering algorithm[J]. Pattern recognition letters, 2003, 24(9/10): 1607–1612. [9] HUNG W L, YANG M S, CHEN Dehua. Parameter se￾lection for suppressed fuzzy c-means with an application to MRI segmentation[J]. Pattern recognition letters, 2006, 27(5): 424–438. [10] SZILÁGYI L, BENYÓ Z, SZILÁGYI S M, et al. MR brain image segmentation using an enhanced fuzzy c￾means algorithm[C]//Proceedings of the 25th Annual In￾ternational Conference of the IEEE Engineering in Medi- [11] 第 4 期 朱占龙，等：融入类贡献抑制因子的灰度级模糊 C 均值图像分割 ·647·