Point Cloud Registration and Modeling Method for Gear Surfaces Based on Laser and Vision Fusion

Xingbang Zhao; Zhengminqing Li; Xiaofeng Yu; Yong Liu; Letian Li

doi:10.3788/LOP241929

Laser & Optoelectronics Progress, Volume. 62, Issue 8, 0815013(2025)

Point Cloud Registration and Modeling Method for Gear Surfaces Based on Laser and Vision Fusion

Xingbang Zhao^1、*, Zhengminqing Li¹, Xiaofeng Yu¹, Yong Liu², and Letian Li¹

¹College of Mechanical & Electrical Engineering, Nanjing University of Aeronautics and Astronautics, Jiangsu 210016, Nanjing , China

²AECC Zhongchuan Transmission Machinery Co., Ltd., Changsha 410000, Hunan , China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(12)

Fig. 1. Line laser measuring coordinate system

Download full size

Fig. 2. Line laser scanning path planning

Download full size

Fig. 3. Rough registration results of tooth surface point cloud. (a)‒(d) Point cloud data of path 1~4; (e)‒(h) point cloud rough registration of path 1~4

Download full size

Fig. 4. Laser measurement platform

Download full size

Fig. 5. Visual measurement platform

Download full size

Fig. 6. Comparison of point cloud coarse registration results. (a) SAC-IA; (b) RANSAC; (c) KD-Super4PCS; (d) ISS-3DSC; (e) ours; (f) SAC-IA+ICP; (g) RANSAC+ICP; (h) KD-Super4PCS+ICP; (i) ISS-3DSC +ICP; (j) ours+ICP

Download full size

Fig. 7. Comparison of point cloud fine registration results. (a) Source data; (b) ICP; (c) KD-ICP; (d) LM-ICP; (e) ours

Download full size

Fig. 8. Running time of precise registration algorithm

Download full size

Fig. 9. Realistic modeling of tooth surface. (a) Geometric three-dimensional reconstruction of tooth surface; (b) texture mapping of tooth surface

Download full size

Table 1. Comparison of root mean square error of fine registration
View table
Table 1. Comparison of root mean square error of fine registration
Point cloud data ICP KD-ICP LM-ICP Ours
1-2 0.24374 0.23186 0.10315 0.09884
1-4 0.25688 0.26841 0.10269 0.09521
2-3 0.23151 0.25213 0.22269 0.10788
4-3 0.31816 0.33855 0.10058 0.10673
Mean value 0.26257 0.27274 0.13228 0.10217

Table 2. Comparison of precise registration time
View table
Table 2. Comparison of precise registration time
Point cloud data ICP KD-ICP LM-ICP Ours
1-2 31.733 22.424 35.321 14.708
1-4 29.715 24.785 36.959 15.072
2-3 32.413 22.636 35.661 14.471
4-3 31.591 25.183 35.158 14.838
Mean value 31.363 23.757 35.775 14.772

Table 3. Image pose solution results

View table

Table 3. Image pose solution results

Parameter	Parameter solution result
Intrinsic parameter	$[\begin{matrix} 2893 & 0 & 1525.3 \\ 0 & 2891.8 & 1122.4 \\ 0 & 0 & 1 \end{matrix}]$
Extrinsic parameter	$[\begin{matrix} 0.9208 & 0.0016 & 0.09182 & - 3.8394 \\ - 0.0179 & - 0.9387 & - 0.2908 & - 40.9365 \\ 0.0097 & 0.1316 & - 0.9913 & 120.8705 \\ 0 & 0 & 0 & 1 \end{matrix}]$
k₁	-0.0057
k₂	-0.0202
p₁	-9.6977×10^-5
p₂	-6.4096×10^-4

Tools

Get Citation

Copy Citation Text

Xingbang Zhao, Zhengminqing Li, Xiaofeng Yu, Yong Liu, Letian Li. Point Cloud Registration and Modeling Method for Gear Surfaces Based on Laser and Vision Fusion[J]. Laser & Optoelectronics Progress, 2025, 62(8): 0815013

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites