面向医学图像分割的蚁群密度峰值聚类
|
摘要
在医学图像分割研究中,针对密度峰值聚类算法(density peaks clustering algorithm,DPC),依靠先验知识给定截断距离d_c且人工选择聚类中心点具有主观随意性等缺陷,提出了一种结合蚁群算法选取密度峰值聚类最优参数的医学图像分割方法.该算法首先利用蚁群算法全局性和鲁棒性的优点,使用图像熵计算信息素来指导蚁群的搜索路径;再使用变量量化表示聚类中心个数,蚁群通过迭代选择最优截断距离d_c和聚类中心,实现了DPC算法的自适应分割并得到了较好的分割效果.仿真实验分析证明了算法的有效性和实用性.
In the medical image segmentation research,a medical image segmentation method based on ant colony algorithm for selecting the optimal parameters of density peaks clustering(DPC)was proposed. For some defects in DPC algorithm,such as cut-off distance d_c was given by DPC algorithm relied on prior knowledge,subjective randomness in cluster centers was selected by manual work. First,the algorithm took advantages of the overall robustness of the ant colony algorithm,used image entropy to calculate pheromone to guide the search path of ant colony. Then it was quantified the number of cluster centers by using variable quantification instead,and ant colony selected the optimal truncation distance d_c and cluster centers by iteration,and realized the adaptive segmentation of DPC algorithm and obtained better results. Simulation experiments proved the effectiveness and practicability of this algorithm.
In the medical image segmentation research,a medical image segmentation method based on ant colony algorithm for selecting the optimal parameters of density peaks clustering(DPC)was proposed. For some defects in DPC algorithm,such as cut-off distance d_c was given by DPC algorithm relied on prior knowledge,subjective randomness in cluster centers was selected by manual work. First,the algorithm took advantages of the overall robustness of the ant colony algorithm,used image entropy to calculate pheromone to guide the search path of ant colony. Then it was quantified the number of cluster centers by using variable quantification instead,and ant colony selected the optimal truncation distance d_c and cluster centers by iteration,and realized the adaptive segmentation of DPC algorithm and obtained better results. Simulation experiments proved the effectiveness and practicability of this algorithm.
引文
[1] SHRIVASTAVA K,GUPTA N,SHARMA N.Medical image segmentation using modified K-means clustering[J].Journal of gerontological nursing,2014,24(1):54.
[2] YADAV H,BANSAL P,SUNKARIA R K.Color dependent K-means clustering for color image segmentation of colored medical images[C]//2015 First International Conference on Next Generation Computing Technologies.Dehardun,India:IEEE,2016:858-862.
[3] ZHOU N,YANG T,ZHANG S.An improved FCM medical image segmentation algorithm based on MMTD[J].Computational and mathematical methods in medicine,2014(1):690349.
[4] 唐文静,许兆新,张小峰.峰值检测FCM算法的医学图像分割[J].智能系统学报,2014(5):584-589.
[5] ZHU H,XU J,HU J,et al.Medical image segmentation using improved affinity propagation[C]//Fifth International Symposium Computational Modeling of Objects Represented in Images.Niagrara Falls NY,USA:Springer,Cham,2016:208-215.
[6] RODRIGUEZ A,LAIO A.Clustering by fast search and find of density peaks[J].Science,2014,344(6191):1492-1496.
[7] 杨光.基于密度峰值聚类的前列腺KVCT图像3D分割[D].西安:西安电子科技大学,2015.
[8] 杨悦,刘向阳.基于局部空间密度峰值的图像分割算法[J].信息技术,2017(9):84-87.
[9] 王泽,张宏军,张睿,等.基于遗传算法与密度及距离计算的聚类方法[J].计算机应用,2015,35(11):3243-3246.
[10] DENG S L,CHEN W D.K-means clustering method based on genetic simulated annealing algorithm[J].Electronic design engineering,2014,22(6):54-56.
[11] 杨志,罗可.一种改进的基于粒子群的聚类算法[J].计算机应用研究,2014,31(9):2597-2599.
[12] ZHANG Z,YUE B,PAN Q,et al.Image segmentation algorithm of fuzzy clustering based on ant colony and adaptive filtering[J].Application of electronic technique,2015,41(4):144-147.
[13] YANG J,TAN W,MA S,et al.Automatic MRI brain tissue extraction algorithm based on three-dimensional gray-scale transformation model[J].Journal of medical imaging and health informatics,2014,4(6):907-911.
[14] JENG J H,HSU S L,CHANG Y.Entropy improvement for fractal image coder[J].International Arab journal of information technology,2012,9(5):403-410.
[15] GUI Y,LIN D,LIANG S,et al.A coarse-to-fine horizon detection method based on between-class variance[C]//Ninth International Symposium on Computational Intelligence and Design.Hangzhou,China:IEEE,2017:376-380.
[2] YADAV H,BANSAL P,SUNKARIA R K.Color dependent K-means clustering for color image segmentation of colored medical images[C]//2015 First International Conference on Next Generation Computing Technologies.Dehardun,India:IEEE,2016:858-862.
[3] ZHOU N,YANG T,ZHANG S.An improved FCM medical image segmentation algorithm based on MMTD[J].Computational and mathematical methods in medicine,2014(1):690349.
[4] 唐文静,许兆新,张小峰.峰值检测FCM算法的医学图像分割[J].智能系统学报,2014(5):584-589.
[5] ZHU H,XU J,HU J,et al.Medical image segmentation using improved affinity propagation[C]//Fifth International Symposium Computational Modeling of Objects Represented in Images.Niagrara Falls NY,USA:Springer,Cham,2016:208-215.
[6] RODRIGUEZ A,LAIO A.Clustering by fast search and find of density peaks[J].Science,2014,344(6191):1492-1496.
[7] 杨光.基于密度峰值聚类的前列腺KVCT图像3D分割[D].西安:西安电子科技大学,2015.
[8] 杨悦,刘向阳.基于局部空间密度峰值的图像分割算法[J].信息技术,2017(9):84-87.
[9] 王泽,张宏军,张睿,等.基于遗传算法与密度及距离计算的聚类方法[J].计算机应用,2015,35(11):3243-3246.
[10] DENG S L,CHEN W D.K-means clustering method based on genetic simulated annealing algorithm[J].Electronic design engineering,2014,22(6):54-56.
[11] 杨志,罗可.一种改进的基于粒子群的聚类算法[J].计算机应用研究,2014,31(9):2597-2599.
[12] ZHANG Z,YUE B,PAN Q,et al.Image segmentation algorithm of fuzzy clustering based on ant colony and adaptive filtering[J].Application of electronic technique,2015,41(4):144-147.
[13] YANG J,TAN W,MA S,et al.Automatic MRI brain tissue extraction algorithm based on three-dimensional gray-scale transformation model[J].Journal of medical imaging and health informatics,2014,4(6):907-911.
[14] JENG J H,HSU S L,CHANG Y.Entropy improvement for fractal image coder[J].International Arab journal of information technology,2012,9(5):403-410.
[15] GUI Y,LIN D,LIANG S,et al.A coarse-to-fine horizon detection method based on between-class variance[C]//Ninth International Symposium on Computational Intelligence and Design.Hangzhou,China:IEEE,2017:376-380.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |