| 改进的多视图点云数据配准方法 |
| |
| 引用本文： | 牛丹华,刘桂华,雷清云,牛振启. 改进的多视图点云数据配准方法[J]. 工具技术, 2012, 46(10): 86-90 |
| |
| 作者姓名： | 牛丹华 刘桂华 雷清云 牛振启 |
| |
| 作者单位： | 1. 西南科技大学
2. 中国北方车辆研究所 |
| |
| 摘 要： | 传统最近点迭代(ICP)算法在进行点云数据配准时,由于待配准的点集数据量很大,每个点云都要遍历一遍,所以时效性不高而且误匹配率大。针对此问题,提出先用Canny边缘检测算子对点云数据进行预处理,以此简化预处理点云的数据量,然后用K-D树搜索数据,最后再用ICP算法进行点云配准,以此来达到加快配准速度。实验证明,该方法灵活实用,简化了待匹配的点云集,能很好的解决传统最近点迭代算法中匹配慢的问题,可以很好的满足工程需要。
|
| 关 键 词： | 点云配准 Canny算子 最近点迭代(ICP) |
Improved Multiview Cloud Data Registration Method |
| |
| Niu Danhua , Liu Guihua , Lei Qingyun , Niu Zhenqi. Improved Multiview Cloud Data Registration Method[J]. Tool Engineering(The Magazine for Cutting & Measuring Engineering), 2012, 46(10): 86-90 |
| |
| Authors: | Niu Danhua Liu Guihua Lei Qingyun Niu Zhenqi |
| |
| Affiliation: | Niu Danhua, Liu Guihua, Lei Qingyun, Niu Zhenqi |
| |
| Abstract: | Because of there are large quantity of points cloud data need to match, so if use the traditional Iterative Closest Point （ICP） algorithm, each points cloud data should be traversed one time, that will waste much time and the false match rate is high . In order to obtain the faster speed of the registration, Canny edge detection operator carl be used to preprocess the data firstly to simplify the pretreatment of points cloud data, then use K - D tree to search points cloud data, last match the points cloud with ICP algorithm . The experimental results prove that the method is flexible, practical, sim- plify pretreatment points cloud ; and it can be a good solution to solve the problem of slow matching with the traditional ICP method. So it will be well to meet the requirement of engineering. |
| |
| Keywords: | points cloud registration Canny operator iterative closest point （ICP） |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
