基于改进的SAC-IA-SICP点云配准方法研究

Research on Point Cloud Registration Method Based on SAC-IA-SICP

为了改善传统ICP算法迭代误差大、配准精度低的问题,本文提出一种基于采样一致性配准算法(Sample Consensus Initial Aligment,SAC-IA)初始匹配与改进迭代最近点(Iterative Closet Point,ICP)精配准相结合的配准方法。首先采用SAC-IA进行初始配准,然后将一种对称的目标函数引入ICP算法,提高ICP算法收敛性,并用于点云精配准。实验结果表明,本文方法的配准精度较ICP算法提升了93.00%,时效性提高了15.20%,表明SAC-IA-SICP配准方法可靠性较高。

In order to solve the problems of large iterative error and low registration accuracy of traditional ICP algorithm,a registration method based on the combination of sample consensus initial aligment(SAC-IA)and improved iterative closest point(ICP)precise registration is proposed in this paper.Firstly,SAC-IA was applied for initial registration,then a symmetrical objective function is introduced into ICP algorithm to improve the convergence of ICP algorithm.Then,the improved ICP algorithm is applied to point cloud precise registration.The experiments results show that the registration accuracy of the proposed method is improved by 93%,and the timeliness is improved by 15.20%compared with the original ICP algorithm,indicating that the registration method based on SAC-IA-SICP algorithm is more reliable.