基于形状的三维图像匹配插值方法  

Matching Interpolation of 3-D Images Based on Shape

作　　者：毛欣炜[1] 尹学松[2] 邹鹏程[2] 张谦[2] 

机构地区：[1]浙江大学建筑工程学院土木系,浙江杭州310027 [2]杭州电子工业学院计算机科学与工程系,浙江杭州310018

出　　处：《计算机仿真》2005年第3期107-110,共4页Computer Simulation

摘　　要：现有的插值方法在进行断层图像插值时,要么不能兼顾灰度和形状的变化,要么计算量太大。为解决这一问题,文中提出一种基于形状的三维图像匹配插值算法。通过对已知两幅断层图像进行门限分割,得到各个密度物质的区域后,利用数学形态学的方法确定出插值图像每个分割区域的轮廓,然后在每个轮廓内运用匹配插值方法得到插值点的值,从而得到整个插值图像。新算法将图像的形状信息和灰度信息较好地结合起来,因此,插值出的图像不仅克服了不同密度物质边界模糊问题,而且也克服了传统的基于形状插值方法的缺点。与线性插值相比,新算法插值出的图像视觉效果好;与小波插值相比,新算法的计算量极大地减少。插值结果可有效地应用于构建三维体模型。In the case of image interpolation for 3D volume models, present methods either lack of the capability of interpolating gray levels and shapes at the same time, or need higher computation cost. In order to solve the problem, the paper introduces a shape-based 3D image matching interpolation algorithm. Firstly, the original images are segmented several regions with the threshold segmentation. Secondly, the contours of every region in the interpolated image are determined with the help of mathematical morphology. Finally, the values of interpolated points in the contour are obtained with the matching interpolation. Further, the whole image is obtained. The interpolated image overcomes not only the boundary blur of the different density matter, but also the shortcomings of the traditional shape-based interpolation algorithms. Compared with linear interpolation, the proposed algorithm greatly improves the quality of image. Moreover, the new algorithm has much lower computation cost compared with wavelet-based interpolation. The interpolation can be effectively used to construct 3D volume models.

关 键 词：三维图像 匹配插值 数学形态学 门限分割 

分 类 号：TP317.4[自动化与计算机技术—计算机软件与理论]