A Speedy Point Cloud Registration Method Based on Region Feature Extraction in Intelligent Driving Scene
Abstract
:1. Introduction
- (1)
- A new point cloud registration algorithm is proposed in this paper, which exhibits high accuracy, real-time performance, and reliability.
- (2)
- The selection of point clouds in key regions makes only a very small number of point clouds available for registration.
- (3)
- The local geometric features of the point cloud are introduced in our method to complete the point cloud registration process under the constraints of the key point cloud.
2. Research Methods
2.1. Algorithm Framework
2.2. Region of Interest Area
3. Point Cloud Coarse Registration
3.1. FPFH Feature Descriptor
3.2. Singular Value Decomposition to Solve the Transformation Matrix
4. Point Cloud Fine Registration
4.1. Extracting Point Cloud Features
4.2. Searching for Matching Point Pairs
4.3. Calculating the Transformation Matrix
5. Experiments
5.1. Object-Level Point Cloud Registration Experiment
5.2. Multi-Condition Registration Experiment
5.3. Multi-Scene Registration Experiment
5.4. Real-Vehicle Registration Experiment
5.5. Metrological Characteristics Analysis of Lidar
6. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Chen, Q.; ** Point Cloud Registration Using Line Features Detection. Remote Sens. 2019, 12, 61. [Google Scholar] [CrossRef]
- Lin, D.; Bannehr, L.; Ulrich, C.; Maas, H.-J. Evaluating Thermal Attribute Map** Strategies for Oblique Airborne Photogrammetric System AOS-Tx8. Remote Sens. 2019, 12, 112. [Google Scholar] [CrossRef]
- Jieun, B.; Junhyeok, P.; Seongjun, C.; Changwon, L. 3D Global Localization in the Underground Mine Environment Using Mobile LiDAR Map** and Point Cloud Registration. Sensors 2022, 22, 2873. [Google Scholar]
- Zhao, J.; Chu, J.; Feng, G.; Jiang, Z.; Shi, W.; Gao, Y. Multimodal Point Cloud Registration Based on Adaptive Feature Region in Radiotherapy Guidance. In Proceedings of the International Conference on Virtual Reality and Visualization, Recife, Brazil, 13–14 November 2020. [Google Scholar]
- Gu, B.; Liu, J.; **ong, H.; Li, T.; Pan, Y. ECPC-ICP: A 6D Vehicle Pose Estimation Method by Fusing the Roadside Lidar Point Cloud and Road Feature. Sensors 2021, 21, 3489. [Google Scholar] [CrossRef]
- He, X.; Pan, S.; Gao, W.; Lu, X. LiDAR-Inertial-GNSS Fusion Positioning System in Urban Environment: Local Accurate Registration and Global Drift-Free. Remote Sens. 2022, 14, 2104. [Google Scholar] [CrossRef]
- Xu, S.; Wang, G.; Wu, D. Research on Internal Positioning Method of Pipeline Robot Based on Data Fusion and Point Cloud Registration; Tsinghua University: Bei**g, China, 2022. [Google Scholar]
- Cheng, L.; Wu, Y.; Chen, S.; Zong, W.; Yuan, Y.; Sun, Y.; Zhuang, Q.; Li, M. A Symmetry-Based Method for LiDAR Point Registration. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2018, 11, 285–299. [Google Scholar] [CrossRef]
- Lu, F.; Chen, G.; Liu, Y.; Zhang, L.; Qu, S.; Liu, S.; Gu, R. HRegNet: A Hierarchical Network for Large-Scale Outdoor LiDAR Point Cloud Registration. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Montreal, QC, Canada, 10–17 October 2021; pp. 16014–16023. [Google Scholar]
- Wang, W.; Zhao, C.; Zhang, H. PR-Alignment: Multidimensional Adaptive Registration Algorithm Based on Practical Application Scenarios. Machines 2023, 11, 254. [Google Scholar] [CrossRef]
- Rusu, R.B.; Blodow, N.; Beetz, M. Fast Point Feature Histograms (FPFH) for 3D registration. In Proceedings of the IEEE International Conference on Robotics & Automation, Kobe, Japan, 12–17 May 2009. [Google Scholar]
- Geiger, A.; Lenz, P.; Urtasun, R. Are we ready for autonomous driving? the kitti vision benchmark suite. In Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition, Providence, RI, USA, 16–21 June 2012; pp. 3354–3361. [Google Scholar]
- Rusu, R.B.; Blodow, N.; Marton, Z.C.; Beetz, M. Aligning point cloud views using persistent feature histograms. In Proceedings of the IEEE/Rsj International Conference on Intelligent Robots and Systems, Nice, France, 22–26 September 2008; pp. 3384–3391. [Google Scholar]
- Eggert, D.W.; Lorusso, A.; Fisher, R.B. Estimating 3-D rigid body transformations: A comparison of four major algorithms. Mach. Vis. Appl. 1997, 9, 272–290. [Google Scholar] [CrossRef]
- Chen, Y.; Medioni, G. Object Modeling by Registration of Multiple Range Images. Int. J. Image Vis. Comput. 1992, 10, 145–155. [Google Scholar] [CrossRef]
Algorithm | Root Mean Square Error/m | Time/s |
---|---|---|
NDT | 0.032 | 29.32 |
TRICP | 0.036 | 22.71 |
NICP | 0.034 | 7.46 |
Ours | 0.030 | 1.53 |
Time/s | |||||||||
---|---|---|---|---|---|---|---|---|---|
Method | Scene1 | Scene2 | Scene3 | Scene4 | Scene5 | Scene6 | Scene7 | Scene8 | Scene9 |
NDT | 13.80 | 12.50 | 12.75 | 15.47 | 18.48 | 14.83 | 19.42 | 16.08 | 14.32 |
TRICP | 12.24 | 10.37 | 14.43 | 11.98 | 15.72 | 11.73 | 7.18 | 13.51 | 12.20 |
NICP | 12.68 | 12.67 | 19.17 | 13.97 | 11.08 | 12.21 | 13.86 | 15.07 | 15.31 |
Ours | 0.63 | 0.38 | 0.66 | 0.49 | 0.74 | 0.59 | 0.62 | 0.89 | 0.90 |
Method | Road/m | Countryside/m | City/m |
---|---|---|---|
NDT | 0.1427 | 0.1136 | 0.1241 |
TRICP | 0.1346 | 0.08100 | 0.08766 |
NICP | 0.1108 | 0.07548 | 0.08101 |
OURS | 0.07231 | 0.06991 | 0.06766 |
Technical Parameter | |||
---|---|---|---|
Principle of distance measurement | time-of-flight measurement | Scanning frequency | 10 Hz, 20 Hz |
Scanning principle | mechanical rotation | Vertical field of view | 40° (−25~+15°) |
Number of threads | 40 | Vertical angular resolution | minimum 0.33° |
Detection distance | 0.3~200 m | Horizontal field of view | 360° |
Measurement accuracy | ±5 cm (0.3~0.5 m) ±2 cm (0.5~200 m) | Horizontal angular resolution | 0.2° (10 Hz) 0.4° (20 Hz) |
Algorithm | Root Mean Square Error/m |
---|---|
NDT | 0.1229 ± 0.03925 |
NICP | 0.1182 ± 0.04029 |
TRICP | 0.1244 ± 0.06505 |
OURS | 0.05996 ± 0.01751 |
Device Model | Measuring Distance | Ranging Accuracy | Horizontal Field of View | Horizontal Angular Resolution | Vertical Field of View | Vertical Angular Resolution |
---|---|---|---|---|---|---|
Pandar40P | 200 m | ±2~±5 cm | 360° | 0.2~0.4° | 40° | 0.33~6° |
Velodyne HDL_64E | 120 m | ±2 cm | 360° | 0.08~0.35° | 26.9° | 0.4° |
Algorithm | Pandar40P | Velodyne HDL-64E | ||
---|---|---|---|---|
Root Square Mean Error/m | Time/s | Root Square Mean Error/m | Time/s | |
NDT | 0.1229 ± 0.03925 | 27.8158 ± 6.2704 | 0.1169 ± 0.03418 | 23.5018 ± 6.4974 |
NICP | 0.1182 ± 0.04029 | 21.1194 ± 9.9502 | 0.09416 ± 0.04526 | 17.4721 ± 9.9084 |
TRICP | 0.1244 ± 0.06505 | 17.8066 ± 10.8586 | 0.1008 ± 0.05135 | 10.5351 ± 6.7416 |
OURS | 0.05996 ± 0.01751 | 0.5739 ± 0.1227 | 0.06971 ± 0.01597 | 0.5873 ± 0.1470 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Yan, D.; Wang, W.; Li, S.; Sun, P.; Duan, W.; Liu, S. A Speedy Point Cloud Registration Method Based on Region Feature Extraction in Intelligent Driving Scene. Sensors 2023, 23, 4505. https://doi.org/10.3390/s23094505
Yan D, Wang W, Li S, Sun P, Duan W, Liu S. A Speedy Point Cloud Registration Method Based on Region Feature Extraction in Intelligent Driving Scene. Sensors. 2023; 23(9):4505. https://doi.org/10.3390/s23094505
Chicago/Turabian StyleYan, Deli, Weiwang Wang, Shaohua Li, Pengyue Sun, Weiqi Duan, and Sixuan Liu. 2023. "A Speedy Point Cloud Registration Method Based on Region Feature Extraction in Intelligent Driving Scene" Sensors 23, no. 9: 4505. https://doi.org/10.3390/s23094505