| 陆军,彭仲涛,夏桂华.点云多法向量邻域特征配准算法[J].光电子激光,2015,26(4):780 |
| 点云多法向量邻域特征配准算法 |
| Point cloud registration algorithm based on neighborhood features of multi-scal e normal vectors |
| 投稿时间：2014-12-24 |
| DOI： |
| 中文关键词: 点云配准 多法向量邻域特征 主成分分析(PCA) 法向量 对应关系 |
| 英文关键词:point cloud registration neighborhood features of multi-scale vector s principal components analysis (PCA) normal vector correspondence |
| 基金项目:黑龙江省自然基金(F201123)、中央高校基本科研业务费专项基金(HEUCFX41304)资助项目 (哈尔滨工程大学自动化学院,黑龙江哈尔滨 150001) |
|
| 摘要点击次数: 963 |
| 全文下载次数: 9 |
| 中文摘要: |
| 针对三维激光扫描点云数据的配准问题,提出了 一种多法向量邻域特征点云配准算法。首先,根据目标点选取不同邻 域半径估算的法向量存在方向偏差,设定约束条件选择关键点,使得初始点云数据量得到精 简；其次,设计了一种依据邻域多 法向量计算的特征描述子,并计算所有关键点的特征向量；然后,依据所求的特征描 述子,使用最小距离与次小距离比值阈 值方法初步获取对应关系,并使用随机采样一致性算法和聚类分选方法进行两次优化,得到 精确的点与点对应关系；最后,使 用奇异值分解法解算刚体变换矩阵,得到配准参数。实验结果表明,由本文设计的关键点选 取、特征描述子提取和对应关系筛选 方法原理简单、稳定可靠、计算速度较快且计算复杂度小,无需进行第二次配准,对实现点 云配准具有实用价值。 |
| 英文摘要: |
| To solve the registration of 3D laser scanning point cloud data,a new method of registration algorithm based on neighborhood features of multi-scale vectors is proposed.Firstly,because the re is error between normal vectors of a point calculated by different neighborhood radii,setting constraint condition can be used to sel ect the key points.Thus,the point cloud data is streamlined.Secondly,a method for extracting point feature information is des igned based on neighborhood eigenvectors and feature descriptor of all key points can be gotten by using this method.Then,by using th e minimum and second distance ratio thresholds to obtain rough corresponding relation and twice optimization methods (random sample con sensus algorithm and clustering sorting method), the exact correspondence between source point and target point cloud can be gott en.Finally,covariance matrix is built and decomposed to get the rigid body transformation matrix.The experimental res ults show that the selection of key points, extraction of point feature information and determination of correspondence of t he new method have simple theory,stable performance,high calculation speed and low computational complexity,and it has practical significance to th e realization of point cloud registration. |
| 查看全文 下载PDF阅读器 |
| 关闭 |
|
|
|
|
|