摘要
针对图形模糊聚类算法缺乏噪声抑制能力的不足,提出基于鲁棒距离的自适应图形模糊聚类分割算法.首先,将邻域像素灰度信息嵌入图形模糊聚类目标函数,得到鲁棒图形模糊聚类分割算法.然后,利用鲁棒距离代替鲁棒图形模糊聚类目标函数中的平方欧氏距离,并对该鲁棒聚类中正则因子采用当前样本与邻域信息均值之偏差进行自适应调节.最后,利用拉格朗日乘子法获得自适应鲁棒图形模糊聚类迭代表达式.灰度图像及其噪声干扰图像的分割测试结果表明:该分割算法相比图形模糊聚类算法、鲁棒图形模糊聚类算法以及现有的鲁棒模糊聚类算法等具有更强的分割能力和抑制噪声的能力.
For picture fuzzy clustering without ability of suppressing noise,an adaptive picture fuzzy clustering segmentation algorithm based on robust distance was proposed.Firstly,the gray level information of neighborhood pixels was embedded into the objective function of picture fuzzy clustering,and the robust image segmentation algorithm based on picture fuzzy clustering was obtained.Secondly,the squared Euclidean distance in the objective function of the robust picture fuzzy clustering was replaced by the robust distance of absolute function,and the regular factor in the robust picture fuzzy clustering was adaptively adjusted by the deviation of the current clustering pixel and the mean of its neighborhood information.Finally,the iterative expression of the adaptive robust picture fuzzy clustering based on the robust distance was obtained by using Lagrange multiplier method.Some segmentation results of gray images and their noised images show that the proposed segmentation algorithm has better segmentation performance and stronger ability of suppressing noise than picture fuzzy clustering algorithm,robust picture fuzzy clustering algorithm and existing robust fuzzy clustering algorithm.
引文
[1]SELVARAJ D,DHANASEKARAN R.MRI brain image segmentation techniques-a review[J].Indian Journal of Computer Science and Engineering,2013,4(5):364-381.
[2]余先川,贺辉,胡丹,等.基于区间值模糊C-均值算法的土地覆盖分类[J].中国科学:地球科学,2014,44(9):2022-2029.
[3]CHAIRA T.A novel intuitionistic fuzzy C means clustering algorithm and its application to medical images[J].Applied Soft Computing,2011,11(2):1711-1717.
[4]KAUR P.Intuitionistic fuzzy sets based credibilistic fuzzy C-means clustering for medical image segmentation tion[J].International Journal of Information Technology,2017,9(4):345-351.
[5]KUMAR D,VERMA H,MEHRA A,et al.A modified intuitionistic fuzzy C-means clustering approach to segment human brain MRI image[J/OL].[2018-05-18].https://doi.org/10.1007/s11042-01010-5954-0.
[6]CUONG B C.Picture fuzzy sets[J].Computer Science Cybernetics,2014,30(4):409-420.
[7]SON L H.DPFCM:A novel distributed picture fuzzy clustering method on picture fuzzy sets[J].Expert Systems with Applications,2015,42:51-66.
[8]THONG P H,SON L H.Picture fuzzy clustering:a new computational intelligence method[J].Soft Computing,2016,20:3549-3562.
[9]ARUHA KUMAR S V,HARISH B S,MANJUNATHARADHYA V N.A picture fuzzy clustering approach for brain tumor segmentation[C]//Proc of 2016 Second International Conference on Cognitive Computing and Information Processing.Mysore:IEEE,2016:1-6.
[10]SON L H,THONG P H.Some novel hybrid forecast methods based on picture fuzzy clustering for weather nowcasting from satellite image sequences[J].Applied Intelligence,2017,46(1):1-15.
[11]CHEN S C,ZHANG D Q.Robust image segmentation using FCM with spatial constraints based on new kernelinduced distance measure[J].IEEE Transactions on Systems,Man,and Cybernetics,2004,34(4):1907-1916.
[12]KOUNDAL D,SHARMA B,GANDOTRA E.Spatial intuitionistic fuzzy set based image segmentation[J].Ima-ging Medicine,2017,9(4):95-101.
[13]GARCIA ESCUDERO L A,GORDALIZA A,CARLOSMATRAN C,et al.A review of robust clustering methods[J].Advanced Data Analysis Classification,2010,4:89-109.
[14]雍龙泉.一致光滑逼近函数及其性质[J].陕西理工大学学报:自然科学版,2018,34(1):74-79.
[15]LAM S Y.Robust clustering methods for pattern classification[D].Hongkong:Library of City University of Hong Kong,2008.
[16]KRINIDIS S,CHATZIS V.A robust fuzzy local information C-means clustering algorithm[J].IEEE Transactions on Image Processing,2010,19(5):1328-1337.