论文信息 - A Fast Weighted Registration Method of 3D Point Cloud Based on Curvature Feature

A Fast Weighted Registration Method of 3D Point Cloud Based on Curvature Feature

In order to realize the fast and accurate registration of 3D point cloud data, a new fast weighted registration method is proposed in this paper. Firstly, using curvature feature, the method samples the original 3D point cloud data to quickly find matching points and remove wrong point pairs. Secondly, by introducing the iterative re-weighted least squares (IRLS) algorithm, the method carries out coarse alignment of the scattered point cloud. Finally, the method presents an improved distance-weighted Iterative Closest Point (ICP) algorithm to achieve fine matching. The experimental results show that the method has good convergence, robustness and accuracy.

Bin Liang | Bing Liu | Xueqian Wang | Houde Liu | Xuehai Gao

[1] Pavel Krsek,et al. The Trimmed Iterative Closest Point algorithm , 2002, Object recognition supported by user interaction for service robots.

[2] Robert C. Bolles,et al. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography , 1981, CACM.

[3] Andrew W. Fitzgibbon. Robust registration of 2D and 3D point sets , 2003, Image Vis. Comput..

[4] Jianmin Dong,et al. An Improvement of Affine Iterative Closest Point Algorithm for Partial Registration , 2016, ICIMCS.

[5] Marc Rioux,et al. Three-dimensional registration using range and intensity information , 1994, Other Conferences.

[6] Per Bergström,et al. Robust registration of point sets using iteratively reweighted least squares , 2014, Computational Optimization and Applications.

[7] Paul J. Besl,et al. A Method for Registration of 3-D Shapes , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[8] Wen Wang,et al. A New Method for Registration of 3D Point Sets with Low Overlapping Ratios , 2015 .

[9] Per Bergström,et al. Robust registration of surfaces using a refined iterative closest point algorithm with a trust region approach , 2016, Numerical Algorithms.

[10] Bin Liang,et al. An Iterative Closest Points Algorithm for Registration of 3D Laser Scanner Point Clouds with Geometric Features , 2017, Sensors.

[11] Robust Registration of 2 D and 3 D Point Sets , 2001 .