首页 | 本学科首页   官方微博 | 高级检索  
     

基于PCL的柴油机缸盖三维特征提取技术
引用本文:南刚雷,张瑞峰,张梓然.基于PCL的柴油机缸盖三维特征提取技术[J].电子测量技术,2016,39(2):80-82.
作者姓名:南刚雷  张瑞峰  张梓然
作者单位:天津大学电子信息工程学院天津300072,天津大学电子信息工程学院天津300072,天津大学电子信息工程学院天津300072
摘    要:为测量某型号柴油机缸盖的三维特征尺寸,通过CCD、线结构光和闭环反馈触发控制子系统扫描测量得到缸盖的三维点云数据。应用点云库(PCL,point cloud library)中的随机采样一致性算法(RANSAC)进行平面拟合,计算点云表面的法向量,完成测量坐标系转换,建立位置基准。采用PCL中的SAMODEL_CYLINDER模型将圆柱特征点和背景点进行分割,采用PCL函数拟合圆柱轴线向量、轴线上某点坐标和圆柱半径来确定圆柱位置,实现特征提取。实验表明采用PCL中的平面拟合函数、圆柱拟合函数具有较高的计算精度。

关 键 词:PCL  RANSAC  三维平面拟合  三维圆柱拟合

Technology about cylinder head's 3D feature extraction based on PCL
Nan Ganglei,Zhang Ruifeng and Zhang Ziran.Technology about cylinder head's 3D feature extraction based on PCL[J].Electronic Measurement Technology,2016,39(2):80-82.
Authors:Nan Ganglei  Zhang Ruifeng and Zhang Ziran
Affiliation:School of Electronic and Information Engineering, Tianjin University, Tianjin 300072, China,School of Electronic and Information Engineering, Tianjin University, Tianjin 300072, China and School of Electronic and Information Engineering, Tianjin University, Tianjin 300072, China
Abstract:To measure the 3D characteristics of a certain diesel engine cylinder head, the whole point cloud of the cylinder head is obtained through a system consisted of CCD, line structured light and a closed loop feedback control trigger device. The normal vector of surface is calculated. The upper plane of the cylinder head is fitted using random sample consensus in the article. Then the coordinate system is converted, which establishes the foundations for the following position reference. Points on the internal surface of cylinders are extracted using SAMODEL_CYLINDER in PCL. The inner axis vector, coordinate of a certain point on the axis and radius can be obtained through PCL. The measurement results are compared with the standard size, the fitted results using PCL have much accuracy.
Keywords:PCL  random sample consensus algorithm  3D plane fitting  3D cylindrical fitting
本文献已被 万方数据 等数据库收录!
点击此处可从《电子测量技术》浏览原始摘要信息
点击此处可从《电子测量技术》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号