频域技术应用于点云配准研究

Fully Automatic Registration of Point Clouds in Frequency-Domain

点云配准技术在国内外各个领域都在不断的发展并且受到广泛关注,不同的配准方法都有其优势与不足。本文针对地面激光扫描获取的多个测站的点云的拼接问题,提出了一种新的配准方法,即在频域中对点云数据进行配准。该方法使用傅里叶变换将点云从空间域变换到频域,利用傅里叶变换可以将旋转、缩放这两个参数与平移参数分离,即可以单独在频域求解旋转和缩放,之后再在此基础上求解平移参数,从而实现对点云的配准。这种方法具有很强的鲁棒性,而且在频域中,不受点云本身噪声点、密度,或者某些结构缺失的影响,且不需要平移的初始估计。最后的实验结果也表明,本文的算法在实现点云配准时与传统的算法相比具有一定的优势。

The technology of point cloud registration are widely concerned in various fields in domestic and abroad, and different registration approaches have their advantages and disadvantage. To the splicing problem of multi station point cloud obtained from terrestrial laser scanning, this paper proposes a new technique for the registration of point cloud, the registration in the frequency domain. The method in this paper changes point cloud from space domain to frequency domain by the Fourier Transform, and decouples the rotational parameters and the scale parameters from the translational parameters. It calculates the parameters of rotation and scale in the frequency domain, then solves the translation parameters, so as to realize the registration of point cloud. This method has good robustness, and no effects from the noise points, density, or some missing structure, and does not need an initial estimate of the translation. Finally, the experimental results show that, the algorithm compared with the traditional algorithm.