Point cloud registration based on improved 3DSC

Yun-tao ZHAO; Jia-xiang QI; Wei-gang LI; Lei GAN

doi:10.37188/CJLCD.2022-0156

Chinese Journal of Liquid Crystals and Displays, Volume. 37, Issue 12, 1590(2022)

Point cloud registration based on improved 3DSC

Yun-tao ZHAO^1,2, Jia-xiang QI^2、*, Wei-gang LI^1,2, and Lei GAN²

¹Engineering Research Center for Metallurgical Automation and Detecting Technology of Ministry of Education，Wuhan University of Science and Technology，Wuhan 430081，China

²College of Information Science and Engineering，Wuhan University of Science and Technology，Wuhan 430081，China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(11)

Fig. 1. 2DSC extends to 3DSC

Fig. 2. 3DSC meshing

Fig. 3. Downsampling of 3DSC contours

Download full size

View in Article

Fig. 4. Flow chart of improved 3DSC-ICP registration algorithm

Download full size

View in Article

Fig. 5. Downsampling of bunny-shaped contour point cloud

Download full size

View in Article

Fig. 6. Downsampling of flowerpot shape contour point cloud

Download full size

View in Article

Fig. 7. Registration result of bunny

Download full size

View in Article

Fig. 8. Registration result of flowerpot

Download full size

View in Article

Table 1. Comparison of data preprocessing
View table
View in Article
Table 1. Comparison of data preprocessing
点云名称点云数量/个配准方法采样轮廓点云数量/个
斯坦福兔子 35 947 传统3DSC 2 193
本文算法 549
花盆 74 994 传统3DSC 2 045
本文算法 494

Table 2. Comparison of registration time and registration accuracy

View table

View in Article

Table 2. Comparison of registration time and registration accuracy

点云名称	点云数量/个	配准方法	配准时间/s	配准精度/m
斯坦福兔子	35 947	FPFH-ICP	7.365	6.180 42e-05
		PFH-ICP	28.577	5.739 18e-05
		传统3DSC	151.676	3.855 19e-05
		本文算法	28.181	2.253 55e-05
花盆	74 994	FPFH-ICP	3.365	6.381 79e-05
		PFH-ICP	9.635	4.020 3e-05
		传统3DSC	146.236	2.500 94e-05
		本文算法	20.856	9.969 02e-06

Table 3. Comparison of registration errors

View table

View in Article

Table 3. Comparison of registration errors

点云名称	配准方法	平移误差			旋转误差
点云名称	配准方法	X轴	Y轴	Z轴	X轴	Y轴	Z轴
斯坦福兔子	FPFH-ICP	0.028 366 7	-0.297 216	-0.165 49	0.002 447 54	0.413 011	-0.715 686
	PFH-ICP	0.041 312 4	-0.299 695	-0.155 706	0.001 816 12	0.444 923	-0.769 857
	传统3DSC	0.029 028 6	-0.295 562	-0.152 352	0.001 876 67	0.509 763	-0.676 755
	本文算法	0.036 269 2	-0.298 597 3	-0.141 056	0.001 248 98	0.484 872	-0.678 382
花盆	FPFH-ICP	0.002 582 66	-0.312 621	-0.199 191	0.665 39	0.839 475	-0.320 632
	PFH-ICP	0.001 026 71	-0.299 358	-0.199 395	0.659 961	0.839 876	-0.318 798
	传统3DSC	0.001 361 96	-0.296 861	-0.196 721	0.644 82	0.825 733	-0.310 702
	本文算法	0.001 017 62	-0.299 208	-0.198 562	0.641 252	0.820 341	-0.317 877

Tools

Get Citation

Copy Citation Text

Yun-tao ZHAO, Jia-xiang QI, Wei-gang LI, Lei GAN. Point cloud registration based on improved 3DSC[J]. Chinese Journal of Liquid Crystals and Displays, 2022, 37(12): 1590

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Research Articles

Received: May. 6, 2022

Accepted: --

Published Online: Nov. 30, 2022

The Author Email: Jia-xiang QI (q68yifeng@163.com)

DOI:10.37188/CJLCD.2022-0156

Topics

Table 1. Comparison of data preprocessing

Table 1. Comparison of data preprocessing

Table 2. Comparison of registration time and registration accuracy

Table 2. Comparison of registration time and registration accuracy

Table 3. Comparison of registration errors

Table 3. Comparison of registration errors