结合非局部信息的模糊聚类脑MR图像分割
|
摘要
为降低噪声以及偏移场的影响,提出一种基于非局部空间信息的FCM模型。引入非局部信息,克服传统的空间信息仅依赖邻域灰度信息,导致精度不高的缺点,使其在降低噪声影响的同时保持细长拓扑结构区域信息;利用多元高斯分布模型对图像灰度分布进行拟合,构造距离函数,降低传统欧式距离导致鲁棒性不足的影响;利用基函数的线性组合对偏移场进行拟合,将偏移场参数化并耦合到FCM框架下,降低灰度不均匀对分割的影响。实验结果表明,该模型可以得到更精确的分割结果。
To reduce the impact of noise and bias field,a FCM model based on non-local spatial information was proposed.The non-local information was integrated into the model,reducing the impact of noise as well as keeping the image structures.The image gray scale distribution was fitted using the multivariate Gaussian distribution and the distance function was constructed to reduce the effects of lacking robustness caused by the traditional Euclidean distance.To overcome the impact of intensity inhomogeneity,the bias field was approximated at the pixel-by-pixel level by using a linear combination of basis functions,and parameterized using the coefficients of the basis functions.Experimental results of the brain MR images show that the proposed method can obtain more accurate results when segmenting images with noise and bias field.
To reduce the impact of noise and bias field,a FCM model based on non-local spatial information was proposed.The non-local information was integrated into the model,reducing the impact of noise as well as keeping the image structures.The image gray scale distribution was fitted using the multivariate Gaussian distribution and the distance function was constructed to reduce the effects of lacking robustness caused by the traditional Euclidean distance.To overcome the impact of intensity inhomogeneity,the bias field was approximated at the pixel-by-pixel level by using a linear combination of basis functions,and parameterized using the coefficients of the basis functions.Experimental results of the brain MR images show that the proposed method can obtain more accurate results when segmenting images with noise and bias field.
引文
[1]Ahmed MN,Yamany SM,Mohamed N.A modified fuzzy cmeans algorithm for bias field estimation and segmentation of MRI data[J].IEEE Trans on Medical Imaging,2012,21(3):193-199.
[2]Gong MG,Liang Y,Shi J.Fuzzy c-means clustering with local information and kernel metric for image segmentation[J].IEEE Transaction on Image Processing,2013,22(2):573-584.
[3]Pham DL,Prince JL.An adaptive fuzzy c-means algorithm for image segmentation in the presence of intensity inhomogeneities[J].IEEE Pattern Recognition Letters,1999,20(1):57-68.
[4]Li CM,Xu CY,Anderson AW,et al.MRI tissue classification and bias field estimation based on coherent local intensity clustering:A unified energy minimization framework[C]//Proceedings of the 21st International Conference on Information Processing in Medical Imaging.Heidelberg,2009:288-299.
[5]Zhang H,Jonathan QM Wu,Nguyen TM.A robust fuzzy algorithm based on Student’s t-distribution and mean template for image segmentation application[J].IEEE Signal Process Lett,2013,20(2):117-120.
[6]Nguyen T,Wu Q.Fast and robust spatially constrained Gaussian mixture model for image segmentation[J].IEEE Trans Circuits Syst Video Technol,2013,23(1):621-635.
[7]Zexuan Ji,Jinyao Liu,Guo Cao,et al.Chenobust spatially constrained fuzzy c-means algorithm for brain MR image segmentation[J].Pattern Recogn,2014,47(7):2454-2466.
[8]Yousefi S,Azmi R,Zahedi M.Brain tissue segmentation in MR images based on a hybrid of MRF and social algorithms[J].Medical Image Analysis,2012,16(4):840-848.
[9]Roche A,Ribes D,Bach-Cuadra M,et al.On the convergence of EM-like algorithms for image segmentation using Markov random fields[J].Medical Image Analysis,2011,16(6):830-839.
[10]Grewenig S,Zimmer S,Weickert J.Rotationally invariant similarity measures for nonlocal image denoising[J].Journal of Visual Communication and Image Representation,2011,22(2):117-130.
[11]Caldairou B,Passat N,Habas PA,et al.A non-local fuzzy segmentation method:Application to brain MRI[J].Pattern Recogn,2011,44(9):1916-1927.
[12]Li C,Gatenby C,Wang L,et al.Arobust parametric method for bias field estimation and segmentation of MR images[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2009:218-223.
[2]Gong MG,Liang Y,Shi J.Fuzzy c-means clustering with local information and kernel metric for image segmentation[J].IEEE Transaction on Image Processing,2013,22(2):573-584.
[3]Pham DL,Prince JL.An adaptive fuzzy c-means algorithm for image segmentation in the presence of intensity inhomogeneities[J].IEEE Pattern Recognition Letters,1999,20(1):57-68.
[4]Li CM,Xu CY,Anderson AW,et al.MRI tissue classification and bias field estimation based on coherent local intensity clustering:A unified energy minimization framework[C]//Proceedings of the 21st International Conference on Information Processing in Medical Imaging.Heidelberg,2009:288-299.
[5]Zhang H,Jonathan QM Wu,Nguyen TM.A robust fuzzy algorithm based on Student’s t-distribution and mean template for image segmentation application[J].IEEE Signal Process Lett,2013,20(2):117-120.
[6]Nguyen T,Wu Q.Fast and robust spatially constrained Gaussian mixture model for image segmentation[J].IEEE Trans Circuits Syst Video Technol,2013,23(1):621-635.
[7]Zexuan Ji,Jinyao Liu,Guo Cao,et al.Chenobust spatially constrained fuzzy c-means algorithm for brain MR image segmentation[J].Pattern Recogn,2014,47(7):2454-2466.
[8]Yousefi S,Azmi R,Zahedi M.Brain tissue segmentation in MR images based on a hybrid of MRF and social algorithms[J].Medical Image Analysis,2012,16(4):840-848.
[9]Roche A,Ribes D,Bach-Cuadra M,et al.On the convergence of EM-like algorithms for image segmentation using Markov random fields[J].Medical Image Analysis,2011,16(6):830-839.
[10]Grewenig S,Zimmer S,Weickert J.Rotationally invariant similarity measures for nonlocal image denoising[J].Journal of Visual Communication and Image Representation,2011,22(2):117-130.
[11]Caldairou B,Passat N,Habas PA,et al.A non-local fuzzy segmentation method:Application to brain MRI[J].Pattern Recogn,2011,44(9):1916-1927.
[12]Li C,Gatenby C,Wang L,et al.Arobust parametric method for bias field estimation and segmentation of MR images[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition,2009:218-223.