基于改进FCM聚类医学图像配准

2018-01-18 19:56:50陈园刘军华雷超阳
计算技术与自动化 2017年4期

陈园+刘军华+雷超阳

摘 要:ICP和互信息广泛应用于医学图像配准,但存在以下问题:其计算量非常大,耗时长;受初始旋转和平移参数影响较大,图像配准容易造成目标函数陷入局部最优值。该方法通过计算参考图像和浮动图像的质心,获得配准平移初始值;对医学图像坐标进行中心化处理,通过改进的FCM聚类方法把图像坐标聚成2类;把这2个聚类中心拟合成一条直线,可以算出该直线的斜率,得出其倾斜角,从而获得配准旋转初始值。实验结果表明,该方法既可用于单模态图像配准,也可以用于多模态配准。还具有运算量少、图像配准速度较快、计算比较简单、精确度较高等特点,并且解决了图像配准容易陷入局部最优的问题。

关键词:图像配准;fuzzy C-means聚类;迭代最近点;互信息

中图分类号:TP391 文献标志码:A

Medical Image Registration Based on Improved Fuzzy C-means Clustering

CHEN Yuan,LIU Jun-hua,LEI Chao-yang

(Internet Engineering Department, Hunan Post and Telecommunication College, Changsha,Hunan 410015,China)

Abstract:The closest iterative point (ICP) algorithm and the mutual information (MI) technology, as intensity-based and feature-based medical image registration methods respectively, are commonly put into use in medical image registration.But some naturally existing things which restrict the further development need to be faced and be solved.On the one hand, they remain heavily calculation costs and low registration efficiencies.On the other hand, since they seriously depends on whether the initial rotation and translation registration parameters can be exactly extracted, they often traps in the local optimum and even fails to register images.In this paper, we compute the centroids of the reference and floating images by using the image moments to obtain the initial translation values, and use Improved Fuzzy C-means Clustering (FCM) to classify the image coordinates.Before clustering, this proposed method first centralizes the medical image coordinates, creates the two-row coordinate matrix to construct the 2-D sample set to be partitioned into two classes, and computes the slope of a straight line fitted to the two classes, finally derives the rotation angle from solving the arc tangent of the slope and obtains the initial rotation values.Obtains through the experiment,this proposed method can efficiently avoid trapping in the local optimum and is meet the single-mode and multi-mode state image registration.It has a low computational load,a fast registration,a fairly simple implementation and good registration accuracy.

Key words:image registration;FCM clustering;iterative closest points;mutual information

1 引 言

近年來,随着医学影像学的发展,CT、MRI、SPECT和PET等医疗设备的成像质量和图像分辨率都有了很大提高,为临床诊断和治疗提供了可靠的依据。但由于各成像设备成像原理的不同,获得的图像信息也不相同,近年来对于各种医学图像配准技术的研究相当热门。

医学图像配准技术是将CT、MRI、 PET、SPECT等一些医学图像通过计算机相关技术手段实现对于一幅医学图像寻求一种或者一系列的空间变换,使它与另一幅医学图像上的对应点达到空间上的一致[1-2]。近年来,该配准技术得到了专家学者们的高度关注并提出了了很多有效解决问题的方法。其中基于图像特征和基于图像灰度的配准方法[3-5]得到较为的广泛应用,基于图像特征的原理是通过查找图像间共有的并有明显特征的,从而获取变换参数。……

登录APP查看全文