摘要
给出了一种结合最小有向包围盒以及形状分布的识别圆环体及其参数的算法。首先建立基本体素的最小有向包围盒,对体素进行标准化;然后生成其D2形状分布曲线,计算此曲线分别与标准球体形状分布曲线、标准圆形状分布曲线的EMD(Earth Mover’s Distance)值以及相应的副半径;最后通过比较两个副半径的大小来判断此体素是否为圆环体,并由最小有向包围盒的边长给出圆环体的参数。该算法不需要任何交互操作,而且能够识别发生一定外形改变的圆环体,抗噪能力强。
Torus and its parameters recognition algorithm based on shape distributions and minimum oriented bounding box is given.Firstly,the basic voxel is normalized by constructing its minimum oriented bounding box.Secondly,its D2 shape distribution curve is constructed,and the values of Earth Mover＇s Distance （EMD） compared with the corresponding curves of the normal sphere and circle respectively and the corresponding radii of the tube are computed.Finally,the radii are compared and the pa- rameters can be obtained by the bounding box.This algorithm need not interactive manipulation and has a strong antinoise ability to recognize the basic voxel with small changes of shape.
出处
《计算机工程与应用》
CSCD
北大核心
2009年第34期10-12,16,共4页
Computer Engineering and Applications
基金
国家自然科学基金No.60573177
陕西省教育厅专项基金项目(No.09JK491)~~
关键词
逆向工程
体素识别
圆环体
形状分布
reverse engineering
feature recognition
torus
shape distribution
