Point Cloud Registration Algorithm Fusing of Super 4PCS and ICP Based on the Key Points

A point cloud registration algorithm fusing of Super 4PCS and ICP based on the key point is proposed to solve the problem that the traditional Super 4PCS algorithm is time consuming and has poor registration accuracy for point clouds with low-overlap region. Firstly, by using the voxel grid method, point cloud is down-sampled to reduce the amount of the computation data. In order to reduce the search range of consistent four-point sets, keypoints are extracted by using ISS(Intrinsic Shape Signature) method. Then the optimal consistency four-point sets is obtained by Super4PCS based on extracted keypoints. We use each point in this four-point sets as the center to establish a neighborhood ball, and the overlapping regions is obtained by calculating the intersection of the neighborhood balls. Finally, registration is performed by using ICP within obtained overlapping regions. The experimental results show that the proposed method can improve the registration speed while improve the registration accuracy.

[1]  Yu Zhong,et al.  Intrinsic shape signatures: A shape descriptor for 3D object recognition , 2009, 2009 IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops.

[2]  Zeming Lin,et al.  An automatic registration algorithm for the scattered point clouds based on the curvature feature , 2013 .

[3]  Wen Qu,et al.  Boundary points based scale invariant 3D point feature , 2017, J. Vis. Commun. Image Represent..

[4]  N. Mitra,et al.  4-points congruent sets for robust pairwise surface registration , 2008, SIGGRAPH 2008.

[5]  Daniel Cohen-Or,et al.  4-points congruent sets for robust pairwise surface registration , 2008, ACM Trans. Graph..

[6]  Díbio Leandro Borges,et al.  A dynamic approach for approximate pairwise alignment based on 4-points congruence sets of 3D points , 2011, 2011 18th IEEE International Conference on Image Processing.

[7]  Carolina Raposo,et al.  Using 2 point+normal sets for fast registration of point clouds with small overlap , 2017, 2017 IEEE International Conference on Robotics and Automation (ICRA).

[8]  Leonidas J. Guibas,et al.  Estimating surface normals in noisy point cloud data , 2004, Int. J. Comput. Geom. Appl..

[9]  Niloy J. Mitra,et al.  Super4PCS: Fast Global Pointcloud Registration via Smart Indexing , 2019 .

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

[11]  Andrew E. Johnson,et al.  Using Spin Images for Efficient Object Recognition in Cluttered 3D Scenes , 1999, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  Andrew E. Johnson,et al.  Surface matching for object recognition in complex three-dimensional scenes , 1998, Image Vis. Comput..

[13]  Hui Chen,et al.  3D free-form object recognition in range images using local surface patches , 2004, Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004..

[14]  Zhiguo Cao,et al.  A fast and robust local descriptor for 3D point cloud registration , 2016, Inf. Sci..