Lidar Target Point Cloud Alignment Based on Improved Neighborhood Curvature with Iteration Closest Point Algorithm

Yanhong Li; Jianguo Yan; Xiaoyan Wang

doi:10.3788/LOP212521

Journals >Laser & Optoelectronics Progress >Volume 60 >Issue 2 >Page 0228008 > Article

Laser & Optoelectronics Progress
Vol. 60, Issue 2, 0228008 (2023)

Lidar Target Point Cloud Alignment Based on Improved Neighborhood Curvature with Iteration Closest Point Algorithm

Yanhong Li^1,2,*, Jianguo Yan², and Xiaoyan Wang³

Author Affiliations

¹School of Physics & Electronic Engineering, Xianyang Normal University, Xianyang 712000, Shaanxi, China

²School of Automation, Northwestern Polytechnical University, Xi'an 710072, Shaanxi, China

³School of Electrical and Mechanical Engineering, Xi'an University of Architecture and Technology, Xi'an 710055, Shaanxi, China

show less

DOI: 10.3788/LOP212521 Cite this Article Set citation alerts

Yanhong Li, Jianguo Yan, Xiaoyan Wang. Lidar Target Point Cloud Alignment Based on Improved Neighborhood Curvature with Iteration Closest Point Algorithm[J]. Laser & Optoelectronics Progress, 2023, 60(2): 0228008 Copy Citation Text

show less

Registration results of Bunny in different algorithms. (a) Bunny origin cloud; (b) PCA coarse registration; (c) NDT algorithm; (d) ICP algorithm; (e) algorithm in reference [12]; (f) proposed improved ICP algorithm

Fig. 1. Registration results of Bunny in different algorithms. (a) Bunny origin cloud; (b) PCA coarse registration; (c) NDT algorithm; (d) ICP algorithm; (e) algorithm in reference [12]; (f) proposed improved ICP algorithm

Download full size

Registration results of Airplane in different algorithms. (a) Airplane origin cloud; (b) PCA coarse registration; (c) NDT algorithm; (d) ICP algorithm; (e) algorithm in reference [12]; (f) proposed improved ICP algorithm

Fig. 2. Registration results of Airplane in different algorithms. (a) Airplane origin cloud; (b) PCA coarse registration; (c) NDT algorithm; (d) ICP algorithm; (e) algorithm in reference [12]; (f) proposed improved ICP algorithm

Download full size

Fig. 3. Registration results of Land in different algorithms. (a) Land origin cloud; (b) PCA coarse registration; (c) NDT algorithm; (d) ICP algorithm; (e) algorithm in reference [12]; (f) proposed improved ICP algorithm

Download full size

Parameter	$ε_{0}$	$τ$	$e_{0}$
Value	0.025	50	10^-8

Table 1. Point cloud registration parameter setting

Method	Bunny		Airplane
Method	Ave-matching error /（10^-6 m）	Time-consuming /s	Ave-matching error /（10^-6 m）	Time-consuming /s
NDT	16.26	89.6	43.92	180.5
ICP	6.82	65.5	20.85	110.6
Reference ［12］	5.13	32.2	14.38	69.2
Proposed improved ICP	4.88	20.1	13.61	58.4