摘要
针对点云配准算法中的收敛速度以及收敛区间与配准精度之间的矛盾,提出一种由粗到细的点云配准算法。粗配准采用一种变尺度点云配准算法,解决收敛区间与配准精度之间的矛盾;细配准采用一种改进的迭代最近点(ICP)算法,通过设置旋转角约束和动态迭代系数,解决由旋转角变化过大引起的配准效果不佳的问题,并可大幅提高算法的迭代收敛速度。实验结果表明:提出的由粗到细的点云配准算法具有较高的配准精度和速度,是一种有效的点云配准算法。
Aiming at conflict between convergence rate,convergence interval and registration precision of point cloud registration algorithm,a point cloud registration algorithm from coarse to fine is proposed. In coarse registration,a variable scale point cloud registration algorithm is used to solve the conflict between convergence interval and registration precision. In fine registration process,an improved iterative closest point（ ICP） algorithm is used by setting rotation angle constraint and dynamic iterative coefficient in order to solve bad registration effect brought by over large changes of rotation angles and increase the iterative convergence rate of the algorithm greatly. The experimental results show that the point clound registration algorithm from coarse to fine has higher registration precision and rate,it is an effective point cloud registration algorithm.
作者
赵夫群
ZHAO Fu-qun(School of Education Science,Xianyang Normal University,Xianyang 712000,China;School of Information Science and Technology,Northwest University,Xi＇an 710127,China)
出处
《传感器与微系统》
CSCD
2018年第10期143-146,共4页
Transducer and Microsystem Technologies
基金
国家自然科学基金资助项目(61373117)
陕西省社科界重大理论与现实问题研究项目(2017C054)
关键词
点云配准
迭代最近点
变尺度
旋转角约束
动态迭代系数
point cloud registration
iterative closest point （ICP）
variable scale
rotation angle constraint
dynamic iterative coefficient
