图像分割的EMKPFC算法

南京大学学报(自然科学版) ›› 2017, Vol. 53 ›› Issue (3): 569–.

图像分割的EMKPFC算法

唐益明1，2*，赵跟陆1，2，任福继1，2，丰刚永1，2，胡相慧1，2

出版日期:2017-05-30 发布日期:2017-05-30
作者简介: 1.合肥工业大学情感计算与先进智能机器安徽省重点实验室，合肥，230009；
2．合肥工业大学计算机与信息学院，合肥，230009
基金资助:
基金项目：国家自然科学基金(61673156, 61432004)，中国博士后科学基金( 2012M521218，2014T70585)，安徽省自然科学基金(1408085MKL15，1508085QF129)，国家“八六三”高技术研究发展计划(2012AA011103)
收稿日期：2016－09－30
*通讯联系人，Email：tym608@163.com

The EMKPFC algorithm of image segmentation

Tang Yiming1，2*，Zhao Genlu1，2，Ren Fuji1，2，Feng Gangyong1，2，Hu Xianghui1，2

Online:2017-05-30 Published:2017-05-30
About author: 1.Anhui Province Key Laboratory of Affective Computing and Advanced Intelligent Machine，Hefei University of Technology，Hefei，230009，China；
2．School of Computer and Information，Hefei University of Technology，Hefei，230009，China

摘要/Abstract

摘要： 现有基于模糊聚类的图像分割算法对噪声敏感，不能妥善地处理图像的灰度特征与邻域像素之间关系．针对该问题，在可能性聚类的基础上融入多核聚类思想，提出了图像分割的EMKPFC算法(Enhanced Multiple Kernel Possibilistic Fuzzy Cmeans algorithms)．该算法可以有效地利用模糊聚类方法以及可能性聚类算法的优点．进一步地，该算法能够规避普通核算法对于核函数选择的不确定性，增加了算法的抗变换性；对于挑选的多种核函数，凭借权重组合能够满足不同图像对于各种核函数的偏好需求，计算出最佳匹配的权重值．在没有任何先验的情况下，不仅可以进行准确的划分，而且还可以做到划分非线性团状样本．通过对于人造图像、真实图像和医学图像的实验结果表明，所提算法比其他相关基于模糊聚类的图像分割算法都具有更好的效果．

Abstract: Current image segmentation algorithm based on fuzzy clustering is sensitive to noise and cannot properly deal with the relationship between the gray features of the image and the neighborhood pixels.To solve this problem，the idea of multiple kernel clustering and the possibility of clustering are introduced to propose the EMKPFC(Enhanced Multiple Kernel Possibilistic Fuzzy Cmeans)algorithm for image segmentation.This algorithm can effectively combine the advantages of the fuzzy clustering method and the possibility clustering algorithm.Furthermore，the algorithm can avoid the uncertainty of the selection of kernel function and increase the ability of antitransformation.For a variety of chosen kernel functions，it can meet the different demand of images for nuclear functions and calculate the best match of the weight value.In the case of the data without any priori，it is not only able to accurately divide the leading data，but also can be done with the nonlinear classification of non group data.The results of experiments on synthetic images，real images and medical images show that the proposed algorithm has better performance than other image segmentation algorithms.

唐益明1，2*，赵跟陆1，2，任福继1，2，丰刚永1，2，胡相慧1，2. 图像分割的EMKPFC算法[J]. 南京大学学报(自然科学版), 2017, 53(3): 569–.

Tang Yiming1，2*，Zhao Genlu1，2，Ren Fuji1，2，Feng Gangyong1，2，Hu Xianghui1，2. The EMKPFC algorithm of image segmentation[J]. Journal of Nanjing University(Natural Sciences), 2017, 53(3): 569–.

参考文献

[1]　Perret B，Cousty J，Tankyevych O，et al.Directed connected operators：Asymmetric hierarchies for image filtering and segmentation.IEEE Transactions on Pattern Analysis and Machine Intelligence，2015，37(6)：1162－1176.
[2] Sikka K，Sinha N，Singh P K，et al.A fully automated algorithm under modified FCM framework for improved brain MR image segmentation.Magnetic Resonance Imaging，2009，27(7)：994－1004.
[3] Zhou J，Chen C L P，Chen L．Maximumentropybased multiple kernel fuzzy Cmeans clustering algorithm.In：Proceedings of the IEEE International Conference on Systems，Man and Cybernetics.San Diego，USA：IEEE，2014：1198－1203.
[4] Nguyen D D，Ngo L T，Pham L T，et al.Towards hybrid clustering approach to data classification：Multiple kernels based intervalvalued fuzzy Cmeans algorithms.Fuzzy Sets and Systems，2015，279：17－39.
[5] Dunn J C．A fuzzy relative of the ISO DATA process and its use in detecting compact wellseparated clusters.Journal of Cybernetics and Systems，1973，3(3)：33－57.
[6] Bezdek J C．Pattern recognition with fuzzy objective function algorithms.New York：Plenum Press，1981，256.
[7] Ahmed M N，Yamany S M，Mohamed N A，et al.A modified fuzzy Cmeans algorithm for bias field estimation and segmentation of MRI data.IEEE Transactions on Medical Imaging，2002，21(3)：193－199.
[8] Szilagyi L，Benyo Z，Szilagyi S M，et al.MR brain image segmentation using an enhanced fuzzy Cmeans algorithm.In：Proceedings of 25th Annual International Conference of IEEE EMBS.Cancun，Mexico：IEEE，2003：17－21.
[9] Chen S C，Zhang D Q．Robust image segmentation using FCM with spatial constraints based on new kernelinduced distance measure.IEEE Transactions on System Man and Cybernetics－Part B，2004，34(4)：1907－1916.
[10] Krinidis S，Chatzis V．A robust fuzzy local information Cmeans clustering algorithm.IEEE Transactions on Image Processing，2010，19(5)：1328－1337.
[11] Krishnapuram R，Keller J M．A possibilistic approach to clustering.IEEE Transaction on Fuzzy Systems，1993，1(2)：98－110.
[12] Zhu L，Chung F L，Wang S T．Generalized fuzzy Cmeans clustering algorithm with improved fuzzy partitions.IEEE Transactions on Systems，Man，and Cybernetics－Part B：Cybernetics，2009，39(3)：578－591.
[13] Zhao F，Liu H Q，Fan J L．A multiobjective spatial fuzzy clustering algorithm for image segmentation.Applied Soft Computing，2015，30：48－57.
[14] Huang H C，Chuang Y Y．Multiple kernel fuzzy clustering.IEEE Transaction on Fuzzy Systems，2012，20(1)：120－134.
[15] Doring C，Timm H，Borgelt C．An extension to possibilistic fuzzy cluster analysis.Fuzzy Sets and Systems，2004，147：3－16.
[16] Tang Y M，Liu X P．Differently implicational universal triple I method of(1，2，2)type.Computers & Mathematics with Applications，2010，59(6)：1965－1984.
[17] Tang Y M，Ren F J．Universal triple I method for fuzzy reasoning and fuzzy controller.Iranian Journal of Fuzzy Systems，2013，10(5)：1－24.
[18] Tang Y M，Yang X Z．Symmetric implicational method of fuzzy reasoning.International Journal of Approximate Reasoning，2013，54(8)：1034－1048.
[19] Tang Y M，Ren F J．Variable differently implicational algorithm of fuzzy inference.Journal of Intelligent and Fuzzy Systems，2015，28(4)：1885－1897.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed