Fusion Segmentation Network Guided by Adaptive Sampling Radius and Channel Attention Mechanism Module for MLS Point Clouds
Source abstract
Abstract
Road high-precision mobile LiDAR measurement point clouds are the digital infrastructures for high-precision maps, autonomous driving, digital twins, etc. High-precision automated semantic segmentation of road point clouds is a crucial research direction. Aiming at the problem of low semantic segmentation accuracy of existing deep learning networks for inhomogeneous sparse point clouds of mobile LiDAR system measurements (MLS), a deep learning method that adaptively adjusts the sampling radius of region groups according to the point clouds density is proposed. We construct a deep learning road point clouds dataset based on a self-developed mobile LiDAR system to train and test road point clouds semantic segmentation. The overall accuracy of the method for road point clouds segmentation is 98.08%, with an overall mIOU of 0.73 and mIOUs of 0.99, 0.983, 0.99, 0.66, and 0.51 for roads, guardrails, signs, streetlights, and lane lines, respectively. The experimental result shows that the method can achieve more accurate segmentation for inhomogeneous sparse road point clouds of mobile LiDAR systems. Compared with the existing methods, the segmentation accuracy is significantly improved.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Remote Sensing and LiDAR Applications · Advanced Neural Network Applications · 3D Surveying and Cultural Heritage
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Daxing District, Beijing 102616, China
Engineering Research Centre of Representative Building and Architectural Heritage Database, Ministry of Education, Beijing 100044, China
Key Laboratory of Modern Urban Surveying and Mapping, National Administration of Surveying, Mapping and Geoinformation, Beijing 100044, China
Beijing Key Laboratory of Building Heritage Fine Reconstruction and Health Monitoring, Beijing 102616, China
School of Economics and Management, Hubei University of Technology, Wuhan 430068, China
Funding
National Natural Science Foundation of China · Beijing Advanced Innovation Center for Future Urban Design Project · Teacher Support Program for Pyramid Talent Training Project of Beijing University of Civil Engineering and Architecture
References (33)
Show source-provided references
1. Urban Geospatial Information Acquisition Mobile Mapping System based on close-range photogrammetry and IGS site calibration10.1080/10095020.2021.1924084 ↗
2. Intelligent holographic mapping for digital twin cities
3. High-precision detection method for large and complex steel structures based on global registration algorithm and automatic point cloud generation10.1016/j.measurement.2020.108765 ↗
4. Zhao, J.H., Wang, Y.R., Cao, Y.E., Guo, M., Huang, X.F., Zhang, R.J., Dou, X.T., Niu, X.Y., Cui, Y.Y., and Wang, J. (2021). The Fusion Strategy of 2D and 3D Information Based on Deep Learning: A Review. Remote Sens., 13.10.3390/rs13204029 ↗
5. Novel Trajectory Optimization Algorithm of Vehicle-borne LiDAR Mobile Measurement System
6. Accurate Calibration of a Self-Developed Vehicle-Borne LiDAR Scanning System10.1155/2021/8816063 ↗
7. Application of lidar technology in the deformation analysis of Yingxian wooden towers
8. Normal cloud model theory-based comprehensive fuzzy assessment of wooden pagoda safety10.1016/j.culher.2022.02.001 ↗
9. Balado, J., Martinez-Sanchez, J., Arias, P., and Novo, A. (2019). Road Environment Semantic Segmentation with Deep Learning from MLS Point Cloud Data. Sensors, 19.10.3390/s19163466 ↗
10. Design of time synchronization device for mobile lidar measurement system with BeiDou navigation timing
11. Qi, C.R., Su, H., Mo, K., and Guibas, L.J. (2017, January 21–26). PointNet: Deep Learning on Point Sets for 3D Classifification and Segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.
12. Wang, P., Li, W., Liu, S., Zhang, Y., Gao, Z., and Ogunbona, P. (2016, January 4–8). Large-scale Continuous Gesture Recognition Using Convolutional Neural Networks. Proceedings of the 2016 23rd International Conference on Pattern Recognition (ICPR), Cancun, Mexico.10.1109/ICPR.2016.7899600 ↗
13. Point cloud deep learning benchmark data set
14. Qi, C.R., Yi, L., Su, H., and Guibas, L.J. (2017). PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space. arXiv.
15. Wu, W., Qi, Z., and Li, F. (2019, January 15–20). PointConv: Deep Convolutional Networks on 3D Point Clouds. Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.10.1109/CVPR.2019.00985 ↗
16. Jiang, M., Wu, Y., Zhao, T., Zhao, Z., and Lu, C. (2018). PointSIFT: A SIFT-like Network Module for 3D Point Cloud Semantic Segmentation. arXiv.10.1109/IGARSS.2019.8900102 ↗
17. Li, Y., Bu, R., Sun, M., Wu, W., Di, X., and Chen, B. (2018). PointCNN: Convolution On X-Transformed Points. arXiv.
18. Bhanu, B., Lee, S., Ho, C.C., and Henderson, T. (1986, January 27–31). Range data processing: Representation of surfaces by edges. Proceedings of the Eighth International Conference on Pattern Recognition, Paris, France.
19. Jiang, X.Y., Bunke, H., and Meier, U. (1996, January 2–4). Fast range image segmentation using high leve segmentation primitives. Proceedings of the 3rd IEEE Workshop on Applications of Compute Vision, Sarasota, FL, USA.
20. Segmentation through variable order surface fitting10.1109/34.3881 ↗
21. Architectural modeling from sparsely scanned range data10.1007/s11263-007-0105-5 ↗
22. Segmentation of airborne data using a slope adaptive filter10.1016/j.isprsjprs.2005.10.005 ↗
23. Zhu, Z.J., Li, X., Xu, J.H., Yuan, J.H., and Tao, J. (2021). Unstructured Road Segmentation Based on Road Boundary Enhancement Point-Cylinder Network Using LiDAR Sensor. Remote Sens., 13.10.3390/rs13030495 ↗
24. Supervoxel-Based Region Growing Segmentation for Point Cloud Data10.1142/S0218001421540070 ↗
25. Wang, Z., Yang, L., Sheng, Y., and Shen, M. (2021). Pole-like Objects Segmentation and Multiscale Classifification-Based Fusion from Mobile Point Clouds in Road Scenes. Remote Sens., 13.10.3390/rs13214382 ↗
26. A Supervoxel Approach to Road Boundary Enhancement From 3-D LiDAR Point Clouds10.1109/LGRS.2020.3037484 ↗
27. Automated extraction of lane markings from mobile LiDAR point clouds based on fuzzy inference10.1016/j.isprsjprs.2019.12.009 ↗
28. Wang, L., Huang, Y., Hou, Y., Zhang, S., and Shan, J. (2019, January 15–20). Graph attention convolution for point cloud semantic segmentation. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.10.1109/CVPR.2019.01054 ↗
29. Yang, J.C., Zhang, Q., Ni, B.B., Li, L.G., Liu, J.X., and Tian, Q. (2019). Modeling Point Clouds with Self-Attention and Gumbel Subset Sampling. arXiv.10.1109/CVPR.2019.00344 ↗
30. Capsule-Based Networks for Road Marking Extraction and Classification from Mobile LiDAR Point Clouds10.1109/TITS.2020.2990120 ↗
31. A Dense Feature Pyramid Network-Based Deep Learning Model for Road Marking Instance Segmentation Using MLS Point Clouds10.1109/TGRS.2020.2996617 ↗
32. 3D lidar point-cloud projection operator and transfer machine learning for effective road surface features detection and segmentation10.1007/s00371-021-02103-8 ↗
33. Multi-Scale Point-Wise Convolutional Neural Networks for 3D Object Segmentation from LiDAR Point Clouds in Large-Scale Environments10.1109/TITS.2019.2961060 ↗