A fast registration method for multi-view point clouds with low overlap in robotic measurement

Abstract

With the rapid advancement of mechanical automation and intelligent processing technology, accurately measuring the surfaces of complex parts has emerged as a significant research challenge. Robotic measurement technology plays a crucial role in facilitating rapid quality inspections during the manufacturing process due to its inherent flexibility. However, the irregular shapes and viewpoint occlusions of complex parts complicate precise measurement. To address these challenges, this work proposes a point cloud registration network for robotic scanning systems and introduces a DBR-Net (Dual-line Registration Network) to overcome the issues of low overlap rates and perspective occlusion that currently impede the registration of certain workpieces. First, feature extraction is performed using a bilinear encoder and multi-level feature interactions of both point-wise and global features. Subsequently, the features are sampled through unanimous voting and fed into the RANSAC (Random Sample Consensus) algorithm for pose estimation, enabling multi-view point cloud registration. Experimental results demonstrate that this method significantly outperforms many existing techniques in terms of feature extraction and registration accuracy, thereby enhancing the overall performance of point cloud registration.

Keywords

Point cloud registration / Feature interaction / Multi-view / Robotic measurement

CRediT authorship contribution statement

Chuangchuang Li: Writing - original draft, Validation, Investigation. Xubin Lin: Writing - review & editing, Methodology. Zhaoyang Liao: Writing - review & editing, Visualization, Methodology, Conceptualization. Hongmin Wu: Writing - review & editing, Visualization. Zhihao Xu: Writing - review & editing, Supervision. Xuefeng Zhou: Writing - review & editing, Supervision, Methodology.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was co-supported by the National Natural Science Foundation of China (U22A20176), Guangdong Basic and Applied Basic Research Foundation, China (2022B1515120078), the Guangdong Basic and Applied Basic Research Foundation, China (2021A1515110898), GDAS’ Project of Science and Technology Development, China (2022GDASZH-2022010108), and the Key Areas R&D Program of Dongguan City, China (20201200300062).

Appendix A. Supplementary data

Supplementary material related to this article can be found online at https://doi.org/10.1016/j.birob.2024.100195.

References

[1] W. Peng, Y. Wang, Z. Miao, M. Feng, Y. Tang, Viewpoints planning for active 3-D reconstruction of profiled blades using estimated occupancy probabilities (EOP), IEEE Trans. Ind. Electron. 68 (5) (2021) 4109-4119.

[2] H. Xie, W. Li, H. Liu, General geometry calibration using arbitrary free-form surface in a vision-based robot system, IEEE Trans. Ind. Electron. 69 (6) (2022) 5994-6003.

[3] J. Li, et al., The real-time local surface model construction method of unknown-model workpieces for robotic polishing, IEEE/ASME Trans. Mechatronics.

[4] Z. Wu, et al., A systematic point cloud edge detection framework for automatic aircraft skin milling, IEEE Trans. Ind. Inform. 20 (1) (2024) 560-572.

[5] Z. Wu, et al., Gravitational discriminative optimization for multiview reconstruction of free-form surface, IEEE/ASME Trans. Mechatronics 28 (6)(2023) 3226-3237.

[6] L. He, et al., GFOICP: Geometric feature optimized iterative closest point for 3-D point cloud registration, IEEE Trans. Geosci. Remote Sens. 61 (2023) 5704217, 1-17.

[7] P.J. Besl, N.D. McKay, A method for registration of 3D shapes, IEEE Trans. Pattern Anal. Mach. Intell. 14 (2) (1992) 239-256.

[8] M.A. Alnagdawi, S.Z. Mohd Hashim, ORB-PC feature-based image reg-istration, in: 2019 IEEE International Conference on Signal and Image Processing Applications, ICSIPA, Kuala Lumpur, Malaysia, 2019, pp. 111-115.

[9] R. Lei, et al., Deep global feature-based template matching for fast multi-modal image registration, in: 2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS, Brussels, Belgium, 2021, pp. 5457-5460.

[10] L. Pan, et al., OCTRexpert: A feature-based 3D registration method for retinal OCT images, IEEE Trans. Image Process. 29 (2020) 3885-3897.

[11] J. Sun, et al., Multi-view SAR image registration based on feature extraction,in:2021 2nd China International SAR Symposium, CISS, Shanghai, China, 2021, pp. 1-8.

[12] W.-J. Lee, S.-J. Oh, Remote sensing image registration using equivariance features, in: 2021 International Conference on Information Networking, ICOIN, Jeju Island, Korea (South), 2021, pp. 776-781.

[13] W. Lu, et al., DeepVCP: An end-to-end deep neural network for point cloud registration,in:2019 IEEE/CVF International Conference on Computer Vision, ICCV, Seoul, Korea (South), 2019, pp. 12-21.

[14] Y. Wang, et al., CCAG: End-to-end point cloud registration, IEEE Robot. Autom. Lett. 9 (1) (2024) 435-442.

[15] Y. Wang, J. Solomon, Deep closest point: Learning representations for point cloud registration,in:2019 IEEE/CVF International Conference on Computer Vision, ICCV, Seoul, Korea (South), 2019, pp. 3522-3531.

[16] X. Huang, et al., Feature-metric registration: A fast semi-supervised ap-proach for robust point cloud registration without correspondences,in:2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR, Seattle, WA, USA, 2020, pp. 11363-11371.

[17] J. Du, et al., ATNet: Unsupervised point cloud registration network in-tegrating adaptive graph convolution and transformer,in:2023 China Automation Congress, CAC, Chongqing, China, 2023, pp. 5202-5207.

[18] P. Kadam, et al., R-PointHop: A green, accurate, and unsupervised point cloud registration method, IEEE Trans. Image Process. 31 (2022) 2710-2725.

[19] Z. Zhang, et al., A representation separation perspective to correspondence-free unsupervised 3-D point cloud registration, IEEE Geosci. Remote Sens. Lett. 19 (2022) 7003005, 1-5.

[20] X. Huang, et al., Unsupervised point cloud registration by learning unified Gaussian mixture models, IEEE Robot. Autom. Lett. 7 (3) (2023) 7028-7035.

[21] Y. Jiang, et al., GTINet: Global topology-aware interactions for unsupervised point cloud registration, IEEE Trans. Circuits Syst. Video Technol. 34 (7)(2024) 6363-6375.

[22] Z. Chen, et al., VK-Net: Category-level point cloud registration with unsupervised rotation invariant keypoints, in: 2021 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP, Toronto, ON, Canada, 2021, pp. 1900-1904.

[23] Z. Huang, et al., PF-Net: Point fractal network for 3D point cloud com-pletion,in:2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR, Seattle, WA, USA, 2020, pp. 7659-7667.

[24] Q. Mei, et al., PACNet: A high-precision point cloud registration network based on deep learning,in:2021 13th International Conference on Wire-less Communications and Signal Processing, WCSP, Changsha, China, 2021, pp. 1-5.

[25] B. Eckart, et al., MLMD: Maximum likelihood mixture decoupling for fast and accurate point cloud registration,in:2015 International Conference on 3D Vision, Lyon, France, 2015, pp. 241-249.

[26] J. Han, et al., An improved RANSAC registration algorithm based on region covariance descriptor,in:2015 Chinese Automation Congress, CAC, Wuhan, 2015, pp. 746-751.