三维图像运动特征配准方法研究与仿真

Research and Simulation of Non-Rigid Registration Method for Three-Dimensional Image Movement Process

在非刚性三维图像运动过程进行配准的研究中,由于运动过程具有随机性,使得三维图像运动特征参数无法在一定的尺度空间内固定,采用传统的配准方法,在运动特征参数发生变化时,由于无法准确提取尺度空间内三维图像运动特征点,导致图像配准误差大、精度低的问题。提出改进SIFT算法的非刚性三维图像的运动过程配准方法。依据非刚性图像的运动过程特征在尺度空间的不变性,提取三维特征点,并融合于双向匹配算法构建非刚性图像特征点间的匹配关系,在将每帧图像的灰度值视作一维离散随机变量,并同时将每帧图像的均衡分布灰度值转换为随机变量的样本集,采用样本估计方法对随机变量间的相互信息进行详细的计算,进而精确的实现了对非刚性三维图像的运动过程配准。实验结果证明,改进SIFT算法的非刚性三维图像运动配准精确度高。

A motion process registration method of non - rigid 3D image based on improved SIFT algorithm is pro- posed. Based on the invariance of the motion process characteristics of non - rigid images in scale space, the 3D fea- ture points are extracted, and they are used to fuse into the bidirectional matching algorithm to construct the matching relationship between the feature points of non - rigid image. Then, the grey value of each frame image is taken as one -dimensional discrete random variable, meanwhile, the equilibrium distribution gray value of each frame image is transformed to the sample set of random variable. The sample estimation method is used to make detailed calculation for the mutual information between random variables, thus the motion process registration of non -rigid 3D image is accurately realized. The experimental results show that the proposed algorithm has high precision.