基于曲率特性的牙齿三维模型匹配方法研究  

Matching method of 3D model based on carvature charaoteristics of teeth

作　　者：许松松[1] 梅雪[1] 耿建平[2] 

机构地区：[1]南京工业大学自动化与电气工程学院,江苏南京210000 [2]南京工业大学高新技术研究院,江苏南京210000

出　　处：《计算机工程与设计》2013年第8期2811-2815,共5页Computer Engineering and Design

基　　金：江苏省高校自然科学研究基金项目(09KJB 510002)

摘　　要：为提高无标志点情况下牙齿锥形束CT与蓝光扫描三维模型匹配精度,提出了依据曲率特性选取对应点集的匹配方法。取基准模型中三点为基准点,以各基准点的曲率值为特征评价标准,提取出待匹配模型中与基准点曲率值相近似的点作为特征点;通过约束条件,从特征点中进一步选取出与基准点对应的匹配点;利用三点平移变换法计算旋转矩阵R及平移矢量t,从而实现基准模型与匹配模型的初匹配。通过实例仿真验证了所提算法的准确性和有效性。In order to improve the matching accuracy of teeth cone beam CT and blue light scan three-dimensional model without index point,a match method based on curvature characteristics selected corresponding point is put forward.First,3 points from reference model are taken as datum points,and the curvature value of each datum points is set as characteristic evaluation standard,points on the match model as feature points whose curvature value is similar to that of the datum points.Furthermore,corresponding matching points are selected to feature points under constraints followed by using three points translational and transformation method calculations rotation matrix R and translation vector t to match reference model and match model.The results of the simulation experiment verify the proposed algorithm accuracy and effectiveness.

关 键 词：三维扫描模型 拓扑重建 特征点提取 曲率 三维模型匹配 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]