Feature Sensitive Three-Dimensional Point Cloud Simplification using Support Vector Regression
Abstract
Contemporary three-dimensional (3D) scanning devices are characterized by high speed and resolution. They provide dense point clouds that contain abundant data about scanned objects and require computationally intensive and time consuming processing. On the other hand, point clouds usually contain a large amount of redundant data that carry little or no additional information about scanned object geometry. To facilitate further analysis and extraction of relevant information from point cloud, as well as faster transfer of data between different computational devices, it is rational to carry out its simplification at an early stage of the processing. However, the reduction of data during simplification has to ensure high level of information contents preservation; simplification has to be feature sensitive. In this paper we propose a method for feature sensitive simplification of 3D point clouds that is based on epsilon insensitive support vector regression (epsilon-SVR). The proposed m...ethod is intended for structured point clouds. It exploits the flatness property of epsilon-SVR for effective recognition of points in high curvature areas of scanned lines. The points from these areas are kept in simplified point cloud along with a reduced number of points from flat areas. In addition, the proposed method effectively detects the points in the vicinity of sharp edges without additional processing. Proposed simplification method is experimentally verified using three real world case studies. To estimate the quality of the simplification, we employ non-uniform rational b-splines fitting to initial and reduced scan lines.
Keywords:
support vector regression / point cloud simplification / 3D scanning / 3D data acquisitionSource:
Tehnički vjesnik, 2019, 26, 4, 985-994Publisher:
- Univ Osijek, Tech Fac, Slavonski Brod
Funding / projects:
- An innovative ecologically based approach to implementation of intelligent manufacturing systems for production of sheet metal parts (RS-MESTD-Technological Development (TD or TR)-35004)
- Research and development of modelling methods and approaches in manufacturing of dental recoveries with the application of modern technologies and computer aided systems (RS-MESTD-Technological Development (TD or TR)-35020)
DOI: 10.17559/TV-20180328175336
ISSN: 1330-3651
WoS: 000477083700014
Scopus: 2-s2.0-85071020481
Collections
Institution/Community
Mašinski fakultetTY - JOUR AU - Marković, Veljko AU - Jakovljević, Živana AU - Miljković, Zoran PY - 2019 UR - https://machinery.mas.bg.ac.rs/handle/123456789/3128 AB - Contemporary three-dimensional (3D) scanning devices are characterized by high speed and resolution. They provide dense point clouds that contain abundant data about scanned objects and require computationally intensive and time consuming processing. On the other hand, point clouds usually contain a large amount of redundant data that carry little or no additional information about scanned object geometry. To facilitate further analysis and extraction of relevant information from point cloud, as well as faster transfer of data between different computational devices, it is rational to carry out its simplification at an early stage of the processing. However, the reduction of data during simplification has to ensure high level of information contents preservation; simplification has to be feature sensitive. In this paper we propose a method for feature sensitive simplification of 3D point clouds that is based on epsilon insensitive support vector regression (epsilon-SVR). The proposed method is intended for structured point clouds. It exploits the flatness property of epsilon-SVR for effective recognition of points in high curvature areas of scanned lines. The points from these areas are kept in simplified point cloud along with a reduced number of points from flat areas. In addition, the proposed method effectively detects the points in the vicinity of sharp edges without additional processing. Proposed simplification method is experimentally verified using three real world case studies. To estimate the quality of the simplification, we employ non-uniform rational b-splines fitting to initial and reduced scan lines. PB - Univ Osijek, Tech Fac, Slavonski Brod T2 - Tehnički vjesnik T1 - Feature Sensitive Three-Dimensional Point Cloud Simplification using Support Vector Regression EP - 994 IS - 4 SP - 985 VL - 26 DO - 10.17559/TV-20180328175336 ER -
@article{ author = "Marković, Veljko and Jakovljević, Živana and Miljković, Zoran", year = "2019", abstract = "Contemporary three-dimensional (3D) scanning devices are characterized by high speed and resolution. They provide dense point clouds that contain abundant data about scanned objects and require computationally intensive and time consuming processing. On the other hand, point clouds usually contain a large amount of redundant data that carry little or no additional information about scanned object geometry. To facilitate further analysis and extraction of relevant information from point cloud, as well as faster transfer of data between different computational devices, it is rational to carry out its simplification at an early stage of the processing. However, the reduction of data during simplification has to ensure high level of information contents preservation; simplification has to be feature sensitive. In this paper we propose a method for feature sensitive simplification of 3D point clouds that is based on epsilon insensitive support vector regression (epsilon-SVR). The proposed method is intended for structured point clouds. It exploits the flatness property of epsilon-SVR for effective recognition of points in high curvature areas of scanned lines. The points from these areas are kept in simplified point cloud along with a reduced number of points from flat areas. In addition, the proposed method effectively detects the points in the vicinity of sharp edges without additional processing. Proposed simplification method is experimentally verified using three real world case studies. To estimate the quality of the simplification, we employ non-uniform rational b-splines fitting to initial and reduced scan lines.", publisher = "Univ Osijek, Tech Fac, Slavonski Brod", journal = "Tehnički vjesnik", title = "Feature Sensitive Three-Dimensional Point Cloud Simplification using Support Vector Regression", pages = "994-985", number = "4", volume = "26", doi = "10.17559/TV-20180328175336" }
Marković, V., Jakovljević, Ž.,& Miljković, Z.. (2019). Feature Sensitive Three-Dimensional Point Cloud Simplification using Support Vector Regression. in Tehnički vjesnik Univ Osijek, Tech Fac, Slavonski Brod., 26(4), 985-994. https://doi.org/10.17559/TV-20180328175336
Marković V, Jakovljević Ž, Miljković Z. Feature Sensitive Three-Dimensional Point Cloud Simplification using Support Vector Regression. in Tehnički vjesnik. 2019;26(4):985-994. doi:10.17559/TV-20180328175336 .
Marković, Veljko, Jakovljević, Živana, Miljković, Zoran, "Feature Sensitive Three-Dimensional Point Cloud Simplification using Support Vector Regression" in Tehnički vjesnik, 26, no. 4 (2019):985-994, https://doi.org/10.17559/TV-20180328175336 . .