Abstract
In this research, we use hash match of invariants under fixed pair length and local compatibility check of positions or normal vectors to improve the efficiency of two-point normal set (2PNS) point cloud registration algorithm. On the one hand, we use the key value formed by the invariants of base point pairs of fixed length to construct and retrieve the hash table to realize the matching of base point pairs in the two point clouds to be registered to speed up the extraction of candidate transformation matrices. On the other hand, the time consumed in the verification phase is reduced by checking the compatibility between the positions or normal vectors of the corresponding points in the specific areas of the two point clouds under the transformation from the candidate matrix. Through these two improvements, the algorithm significantly reduces the time spent in the point cloud registration algorithm.
Funder
National Natural Science Foundation of China
the Natural Science Foundation of the Higher Education Institutions of Jiangsu Province
Cooperation Fund of Yangzhou Government and College
Publisher
Public Library of Science (PLoS)
Reference26 articles.
1. A Method for Registration of 3-D Shapes;PJ Besl;IEEE Transactions on Pattern Analysis and Machine Intelligence,1992
2. Rusinkiewicz S, Levoy M. Efficient variants of the ICP algorithm. In: Proceedings of International Conference on 3-D Digital Imaging and Modeling, 3DIM. IEEE; 2001. p. 145–152.
3. Robust Euclidean alignment of 3D point sets: The trimmed iterative closest point algorithm;D Chetverikov;Image and Vision Computing,2005
4. Park J, Zhou QY, Koltun V. Colored Point Cloud Registration Revisited. In: Proceedings of the IEEE International Conference on Computer Vision. vol. 2017-Octob; 2017. p. 143–152.
5. Point Set Registration With Similarity and Affine Transformations Based on Bidirectional KMPE Loss;Y Yang;IEEE Transactions on Cybernetics,2019