摘要
针对密集点云数据与CAD模型的配准问题,提出了一种基于简化模型曲率计算的配准方法。该方法通过计算简化模型和CAD模型各点的曲率提取出两模型上某一对应的特征面,根据特征面求出三组对应点对并计算坐标变换矩阵;把得到的变换矩阵应用于简化前的原始点云模型实现模型的预配准;最后通过奇异值分解和最近点迭代相结合的算法实现精确配准。实例表明,该方法实现了密集点云数据与CAD模型的配准,并在保证配准精度的前提下提高了配准的速度,从而验证了方法的有效性和实用性。
For the matching problem between dense cloud data and CAD model,a registration method based on curvatures calculation of simplified model was proposed.Feature surfaces were extracted based on the curvature of simplified and CAD model,three corresponding points were searched through feature surfaces,and the transformation matrix was computed.Then the obtained matrix was applied to original cloud data to achieve pre-registration.The last step was accurate registration that achieved through SVD-ICP algorithm.Experimental results show that the method can realize registration between dense cloud data and CAD model,and demonstrate the efficiency and practicality of the method.
出处
《中国机械工程》
EI
CAS
CSCD
北大核心
2011年第14期1699-1703,共5页
China Mechanical Engineering
基金
航空科学基金资助项目(2008ZE53042)
关键词
密集点云数据
模型简化
特征提取
配准
dense cloud data
model simplification
feature extracting
registration
