结合形态学重建与隶属度滤波的FCM分割算法
|
摘要
传统的模糊C均值聚类(FCM)算法对噪声敏感,常引入局部空间信息来提高FCM算法对图像分割的鲁棒性,但局部空间信息的引入往往需要大量迭代局部空间邻域内像素和聚类中心之间的距离,导致计算复杂度高.为此,提出一种基于形态学闭合重建和隶属度滤波的FCM算法.首先引入形态学闭合重建算法,充分利用局部空间信息来优化数据的分布特征,提高算法的抗噪性并保留图像细节;再通过FCM算法对重建图像的灰度直方图进行聚类;最后利用隶属度滤波修正隶属度矩阵以避免大量的迭代计算.在合成图像和医学图像上进行实验的结果表明,该算法不仅取得了较好的分割效果,而且所需的时间更短、对噪声的鲁棒性更强.
The introduction of local spatial information in the traditional fuzzy C-means clustering(FCM)algorithm, which can improve the robustness of FCM algorithm in image segmentation, is a common solution to the sensitivity to noise. However, the introduction requires a large amount of iterations to calculate the distance between the pixels in the local spatial neighborhood and the clustering center, resulting in high computational complexity. Thus, the paper proposed an improved FCM algorithm based on morphological closed reconstruction and membership filtering. First, the morphological closed reconstruction algorithm was introduced to make full use of local spatial information which can optimize the distribution feature of data. In this way, noise immunity of the algorithm can be improved and more details of the image can be preserved. Then we used the FCM algorithm to cluster the gray histogram of the reconstructed image. Finally, we modified the membership partition matrix by using the membership filtering to avoid massive iterations. The experimental results on synthetic images and medical images demonstrate that the proposed algorithm not only achieves better segmentation result, but also takes less time and more robust to noise.
The introduction of local spatial information in the traditional fuzzy C-means clustering(FCM)algorithm, which can improve the robustness of FCM algorithm in image segmentation, is a common solution to the sensitivity to noise. However, the introduction requires a large amount of iterations to calculate the distance between the pixels in the local spatial neighborhood and the clustering center, resulting in high computational complexity. Thus, the paper proposed an improved FCM algorithm based on morphological closed reconstruction and membership filtering. First, the morphological closed reconstruction algorithm was introduced to make full use of local spatial information which can optimize the distribution feature of data. In this way, noise immunity of the algorithm can be improved and more details of the image can be preserved. Then we used the FCM algorithm to cluster the gray histogram of the reconstructed image. Finally, we modified the membership partition matrix by using the membership filtering to avoid massive iterations. The experimental results on synthetic images and medical images demonstrate that the proposed algorithm not only achieves better segmentation result, but also takes less time and more robust to noise.
引文
[1]Fu Qiongying,Yu Xuchu,Zhang Pengqiang,et al.Clustering algorithm combining extreme learning machine for hyperspectral image[J].Journal of Computer-Aided Design&Computer Graphics,2017,29(8):1416-1424(in Chinese)(付琼莹,余旭初,张鹏强,等.结合极限学习机的高光谱影像聚类算法[J].计算机辅助设计与图形学学报,2017,29(8):1416-1424)
[2]Rouhi R,Jafari M,Kasaei S,et al.Benign and malignant breast tumors classification based on region growing and CNN segmentation[J].Expert Systems with Applications,2015,42(3):990-1002
[3]Malpica N,de Solorzano C O,Vaquero J J,et al.Applying watershed algorithms to the segmentation of clustered nuclei[J].Cytometry,1997,28(4):289-297
[4]Sun Wenyan,Dong Enqing,Cao Zhulou,et al.A robust local segementation method based on fuzzy-energy based active contour[J].Acta Automatica Sinica,2017,43(4):611-621(in Chinese)(孙文燕,董恩清,曹祝楼,等.一种基于模糊主动轮廓的鲁棒局部分割方法[J].自动化学报,2017,43(4):611-621)
[5]Li J,Chen H F,Li G,et al.Salient object detection based on meanshift filtering and fusion of colour information[J].IETImage Processing,2015,9(11):977-985
[6]Wang Tao,Ji Zexuan,Sun Quansen.A segmentation algorithm combined with non-local information and graph cut[J].Journal of Computer-Aided Design&Computer Graphics,2015,27(5):783-791(in Chinese)(王涛,纪则轩,孙权森.结合非局部信息与图割的图像分割算法[J].计算机辅助设计与图形学学报,2015,27(5):783-791)
[7]Hu F,Xia G S,Wang Z F,et al.Unsupervised feature learning via spectral clustering of multidimensional patches for remotely sensed scene classification[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2015,8(5):2015-2030
[8]Gu W,Lv Z H,Hao M.Change detection method for remote sensing images based on an improved Markov random field[J].Multimedia Tools and Applications,2017,76(17):17719-17734
[9]Zhu Wei,Qu Jingyi,Wu Renbiao.Straight convolutional neural networks algorithm based on batch normalization for image classification[J].Journal of Computer-Aided Design&Computer Graphics,2017,29(9):1650-1657(in Chinese)(朱威,屈景怡,吴仁彪.结合批归一化的直通卷积神经网络图像分类算法[J].计算机辅助设计与图形学学报,2017,29(9):1650-1657)
[10]Bezdek J C,Ehrlich R,Full W.FCM:the fuzzy C-means clustering algorithm[J].Computers&Geosciences,1984,10(2/3):191-203
[11]Wilson B,Julia D.An experimental analysis of fuzzy C-means and K-means segmentation algorithm for iron detection in brain SWI using matlab[J].International Journal of Computer Applications,2014,104(15):36-38
[12]Ahmed M N,Yamany S M,Mohamed N,et al.A modified fuzzy C-means algorithm for bias field estimation and segmentation of MRI data[J].IEEE Transactions on Medical Imaging,2002,21(3):193-199
[13]Chen S C,Zhang D Q.Robust image segmentation using FCMwith spatial constraints based on new kernel-induced distance measure[J].IEEE Transactions on Systems,Man and Cybernetics,Part B(Cybernetics),2004,34(4):1907-1916
[14]Han H W,Tian L,Miao J Q.A fuzzy clustering algorithm of automatic classification based on EnFCM[J].Advanced Materials Research,2014,989-994:1489-1492
[15]Zuo Liyun,Luo Chengyu,Zuo Youxiang.Paralleled segmentation cluster algorithm based on EnFCM for large-scale image[J].Microcomputer&Its Applications,2015,34(15):55-58(in Chinese)(左利云,罗成煜,左右祥.基于EnFCM的海量图像聚类分割算法的并行研究[J].微型机与应用,2015,34(15):55-58)
[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
[17]Gong M G,Su L Z,Jia M,et al.Fuzzy clustering with a modified MRF energy function for change detection in synthetic aperture radar images[J].IEEE Transactions on Fuzzy Systems,2014,22(1):98-109
[18]Lavanya M,Kannan P M.Lung lesion detection in CT scan images using the fuzzy local information cluster means(FLICM)automatic segmentation algorithm and back propagation network classification[J].Asian Pacific Journal of Cancer Prevention,2017,18(12):3395-3399
[19]Wu Y P,Peng X Q,Ruan K,et al.Improved image segmentation method based on morphological reconstruction[J].Multimedia Tools and Applications,2017,76(19):19781-19793
[20]Liu Yue,Wang Xiaopeng,Wang Jinquan,et al.Watershed algorithm for brain tumor segmentation based on morphological reconstruction and gradient layered modification[J].Application Research of Computers,2015,32(8):2487-2491(in Chinese)(刘岳,王小鹏,王金全,等.基于形态学重建和梯度分层修正的分水岭脑肿瘤分割[J].计算机应用研究,2015,32(8):2487-2491)
[21]Thamaraichelvi B,Govindarajan Y.A complexity reduced FCMbased segmentation technique for brain MRI Image classification[J].Journal of Medical Imaging and Health Informatics,2015,5(2):202-209
[22]Choudhry M S,Kapoor R.Performance analysis of fuzzy C-means clustering methods for MRI image segmentation[J].Procedia Computer Science,2016,89:749-758
[2]Rouhi R,Jafari M,Kasaei S,et al.Benign and malignant breast tumors classification based on region growing and CNN segmentation[J].Expert Systems with Applications,2015,42(3):990-1002
[3]Malpica N,de Solorzano C O,Vaquero J J,et al.Applying watershed algorithms to the segmentation of clustered nuclei[J].Cytometry,1997,28(4):289-297
[4]Sun Wenyan,Dong Enqing,Cao Zhulou,et al.A robust local segementation method based on fuzzy-energy based active contour[J].Acta Automatica Sinica,2017,43(4):611-621(in Chinese)(孙文燕,董恩清,曹祝楼,等.一种基于模糊主动轮廓的鲁棒局部分割方法[J].自动化学报,2017,43(4):611-621)
[5]Li J,Chen H F,Li G,et al.Salient object detection based on meanshift filtering and fusion of colour information[J].IETImage Processing,2015,9(11):977-985
[6]Wang Tao,Ji Zexuan,Sun Quansen.A segmentation algorithm combined with non-local information and graph cut[J].Journal of Computer-Aided Design&Computer Graphics,2015,27(5):783-791(in Chinese)(王涛,纪则轩,孙权森.结合非局部信息与图割的图像分割算法[J].计算机辅助设计与图形学学报,2015,27(5):783-791)
[7]Hu F,Xia G S,Wang Z F,et al.Unsupervised feature learning via spectral clustering of multidimensional patches for remotely sensed scene classification[J].IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2015,8(5):2015-2030
[8]Gu W,Lv Z H,Hao M.Change detection method for remote sensing images based on an improved Markov random field[J].Multimedia Tools and Applications,2017,76(17):17719-17734
[9]Zhu Wei,Qu Jingyi,Wu Renbiao.Straight convolutional neural networks algorithm based on batch normalization for image classification[J].Journal of Computer-Aided Design&Computer Graphics,2017,29(9):1650-1657(in Chinese)(朱威,屈景怡,吴仁彪.结合批归一化的直通卷积神经网络图像分类算法[J].计算机辅助设计与图形学学报,2017,29(9):1650-1657)
[10]Bezdek J C,Ehrlich R,Full W.FCM:the fuzzy C-means clustering algorithm[J].Computers&Geosciences,1984,10(2/3):191-203
[11]Wilson B,Julia D.An experimental analysis of fuzzy C-means and K-means segmentation algorithm for iron detection in brain SWI using matlab[J].International Journal of Computer Applications,2014,104(15):36-38
[12]Ahmed M N,Yamany S M,Mohamed N,et al.A modified fuzzy C-means algorithm for bias field estimation and segmentation of MRI data[J].IEEE Transactions on Medical Imaging,2002,21(3):193-199
[13]Chen S C,Zhang D Q.Robust image segmentation using FCMwith spatial constraints based on new kernel-induced distance measure[J].IEEE Transactions on Systems,Man and Cybernetics,Part B(Cybernetics),2004,34(4):1907-1916
[14]Han H W,Tian L,Miao J Q.A fuzzy clustering algorithm of automatic classification based on EnFCM[J].Advanced Materials Research,2014,989-994:1489-1492
[15]Zuo Liyun,Luo Chengyu,Zuo Youxiang.Paralleled segmentation cluster algorithm based on EnFCM for large-scale image[J].Microcomputer&Its Applications,2015,34(15):55-58(in Chinese)(左利云,罗成煜,左右祥.基于EnFCM的海量图像聚类分割算法的并行研究[J].微型机与应用,2015,34(15):55-58)
[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
[17]Gong M G,Su L Z,Jia M,et al.Fuzzy clustering with a modified MRF energy function for change detection in synthetic aperture radar images[J].IEEE Transactions on Fuzzy Systems,2014,22(1):98-109
[18]Lavanya M,Kannan P M.Lung lesion detection in CT scan images using the fuzzy local information cluster means(FLICM)automatic segmentation algorithm and back propagation network classification[J].Asian Pacific Journal of Cancer Prevention,2017,18(12):3395-3399
[19]Wu Y P,Peng X Q,Ruan K,et al.Improved image segmentation method based on morphological reconstruction[J].Multimedia Tools and Applications,2017,76(19):19781-19793
[20]Liu Yue,Wang Xiaopeng,Wang Jinquan,et al.Watershed algorithm for brain tumor segmentation based on morphological reconstruction and gradient layered modification[J].Application Research of Computers,2015,32(8):2487-2491(in Chinese)(刘岳,王小鹏,王金全,等.基于形态学重建和梯度分层修正的分水岭脑肿瘤分割[J].计算机应用研究,2015,32(8):2487-2491)
[21]Thamaraichelvi B,Govindarajan Y.A complexity reduced FCMbased segmentation technique for brain MRI Image classification[J].Journal of Medical Imaging and Health Informatics,2015,5(2):202-209
[22]Choudhry M S,Kapoor R.Performance analysis of fuzzy C-means clustering methods for MRI image segmentation[J].Procedia Computer Science,2016,89:749-758