Point Cloud Registration Based on Multi-Scale Feature and Point Distance Constraint

Fig. 4. Point cloud registration result of our algorithm. (a) Original point cloud; (b) key points; (c) matching points and corresponding relations; (d) results of registration

Download full size

Fig. 5. Fragment of the Bunny point cloud model. (a) bun000 and bun045; (b) bun090 and bun045

Download full size

Fig. 6. Registration result of the Bunny point cloud. (a) bun000 and bun045; (b) bun090 and bun045

Download full size

Fig. 7. Experimental results of office point cloud registration. (a) Original point cloud; (b) 4PCS; (c) Super-4PCS; (d) our algorithm

Download full size

Table 1. Size of partial segments of point cloud models in The Stanford 3D Scanning Repository dataset

View table

Table 1. Size of partial segments of point cloud models in The Stanford 3D Scanning Repository dataset

Serial number	Point cloud model	Segments of model	Number of points
1	Bunny	bun000	40256
1	Bunny	bun045	40097
2	Armadillo	armadilloSide_60	23404
2	Armadillo	armadilloSide_45	20647
3	Buddha	happySideRight_336	68890
3	Buddha	happySideRight_312	57499
4	Dragon	dragonStandRight_336	43467
4	Dragon	dragonStandRight_288	24573

Table 2. Size of partial point cloud segments of Bunny model
View table
Table 2. Size of partial point cloud segments of Bunny model
Point cloud model Segments of model Number of points
Bunny bun000 40256
bun045 40097
bun090 30379

Table 3. Registration results of different algorithms unit: mm

View table

Table 3. Registration results of different algorithms unit: mm

Serial number	Algorithm	Root mean square distance error
Serial number	Algorithm	bun000 and bun045	bun045 and bun090
1	ICP	0.000900	0.017943
2	4PCS	0.004802	0.010270
3	Super-4PCS	0.003803	0.010086
4	GICP	0.001229	0.010159
5	Ours-CR	0.000984	0.007196
6	Ours-FR	0.000896	0.006465

Table 4. Experimental results of registration algorithm for the point cloud of office environment
View table
Table 4. Experimental results of registration algorithm for the point cloud of office environment
Serial number Algorithm Root mean square distance error /mm
1 4PCS 0.139283
2 Super-4PCS 0.135356
3 ours 0.094396

Tools

Get Citation

Copy Citation Text

Xuchun Zhang, Hongjun Zhou, Jinjin Zheng, Yi Jin. Point Cloud Registration Based on Multi-Scale Feature and Point Distance Constraint[J]. Laser & Optoelectronics Progress, 2021, 58(24): 2415004

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Machine Vision

Received: Jan. 3, 2021

Accepted: Mar. 3, 2021

Published Online: Nov. 29, 2021

The Author Email: Zheng Jinjin (jjzheng@ustc.edu.cn), Jin Yi (jinyi08@ustc.edu.cn)

DOI:10.3788/LOP202158.2415004

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology

Table 1. Size of partial segments of point cloud models in The Stanford 3D Scanning Repository dataset

Table 1. Size of partial segments of point cloud models in The Stanford 3D Scanning Repository dataset

Table 2. Size of partial point cloud segments of Bunny model

Table 2. Size of partial point cloud segments of Bunny model

Table 3. Registration results of different algorithms unit: mm

Table 3. Registration results of different algorithms unit: mm

Table 4. Experimental results of registration algorithm for the point cloud of office environment

Table 4. Experimental results of registration algorithm for the point cloud of office environment