DEEP LEARNING-BASED ROAD SEGMENTATION OF 3D POINT CLOUDS FOR ASSISTING ROAD ALIGNMENT PARAMETERIZATION
The need for transportation infrastructure digitalization is becoming more important, and efficient data collection and processing workflows have to be established and pose a great research challenge. This paper presents a fully automated method for the geometric parametrization of the road alignmen...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2022-05-01
|
| Series: | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLIII-B2-2022/283/2022/isprs-archives-XLIII-B2-2022-283-2022.pdf |
| _version_ | 1818208098649636864 |
|---|---|
| author | M. Soilán H. Tardy D. González-Aguilera |
| author_facet | M. Soilán H. Tardy D. González-Aguilera |
| author_sort | M. Soilán |
| collection | DOAJ |
| description | The need for transportation infrastructure digitalization is becoming more important, and efficient data collection and processing workflows have to be established and pose a great research challenge. This paper presents a fully automated method for the geometric parametrization of the road alignment from 3D point clouds acquired with a low-cost mobile mapping system. It exploits the Point Transformer Deep Learning architecture in order to segment the 3D point cloud in four different classes, which include road markings. Those markings are then used as a reference to extract the alignment trajectory path, classify its geometries (straight lines, circular arcs, and clothoids) and then parametrize it, extracting data to easily generate alignment data that may follow the standard schema of the Industry Foundation Classes (IFC). Both the deep learning architecture and the geometry parametrization process show promising results to develop automatic workflows that extract precise as-built data of the infrastructure from 3D point clouds. |
| first_indexed | 2024-12-12T04:39:25Z |
| format | Article |
| id | doaj.art-3c195ede6b6c4578845f7998b637dd2f |
| institution | Directory Open Access Journal |
| issn | 1682-1750 2194-9034 |
| language | English |
| last_indexed | 2024-12-12T04:39:25Z |
| publishDate | 2022-05-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| spelling | doaj.art-3c195ede6b6c4578845f7998b637dd2f2022-12-22T00:37:51ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342022-05-01XLIII-B2-202228329010.5194/isprs-archives-XLIII-B2-2022-283-2022DEEP LEARNING-BASED ROAD SEGMENTATION OF 3D POINT CLOUDS FOR ASSISTING ROAD ALIGNMENT PARAMETERIZATIONM. Soilán0H. Tardy1D. González-Aguilera2Department of Cartographic and Terrain Engineering, University of Salamanca, Calle Hornos Caleros 50, 05003 Ávila, SpainDepartment of Cartographic and Terrain Engineering, University of Salamanca, Calle Hornos Caleros 50, 05003 Ávila, SpainDepartment of Cartographic and Terrain Engineering, University of Salamanca, Calle Hornos Caleros 50, 05003 Ávila, SpainThe need for transportation infrastructure digitalization is becoming more important, and efficient data collection and processing workflows have to be established and pose a great research challenge. This paper presents a fully automated method for the geometric parametrization of the road alignment from 3D point clouds acquired with a low-cost mobile mapping system. It exploits the Point Transformer Deep Learning architecture in order to segment the 3D point cloud in four different classes, which include road markings. Those markings are then used as a reference to extract the alignment trajectory path, classify its geometries (straight lines, circular arcs, and clothoids) and then parametrize it, extracting data to easily generate alignment data that may follow the standard schema of the Industry Foundation Classes (IFC). Both the deep learning architecture and the geometry parametrization process show promising results to develop automatic workflows that extract precise as-built data of the infrastructure from 3D point clouds.https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLIII-B2-2022/283/2022/isprs-archives-XLIII-B2-2022-283-2022.pdf |
| spellingShingle | M. Soilán H. Tardy D. González-Aguilera DEEP LEARNING-BASED ROAD SEGMENTATION OF 3D POINT CLOUDS FOR ASSISTING ROAD ALIGNMENT PARAMETERIZATION The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| title | DEEP LEARNING-BASED ROAD SEGMENTATION OF 3D POINT CLOUDS FOR ASSISTING ROAD ALIGNMENT PARAMETERIZATION |
| title_full | DEEP LEARNING-BASED ROAD SEGMENTATION OF 3D POINT CLOUDS FOR ASSISTING ROAD ALIGNMENT PARAMETERIZATION |
| title_fullStr | DEEP LEARNING-BASED ROAD SEGMENTATION OF 3D POINT CLOUDS FOR ASSISTING ROAD ALIGNMENT PARAMETERIZATION |
| title_full_unstemmed | DEEP LEARNING-BASED ROAD SEGMENTATION OF 3D POINT CLOUDS FOR ASSISTING ROAD ALIGNMENT PARAMETERIZATION |
| title_short | DEEP LEARNING-BASED ROAD SEGMENTATION OF 3D POINT CLOUDS FOR ASSISTING ROAD ALIGNMENT PARAMETERIZATION |
| title_sort | deep learning based road segmentation of 3d point clouds for assisting road alignment parameterization |
| url | https://www.int-arch-photogramm-remote-sens-spatial-inf-sci.net/XLIII-B2-2022/283/2022/isprs-archives-XLIII-B2-2022-283-2022.pdf |
| work_keys_str_mv | AT msoilan deeplearningbasedroadsegmentationof3dpointcloudsforassistingroadalignmentparameterization AT htardy deeplearningbasedroadsegmentationof3dpointcloudsforassistingroadalignmentparameterization AT dgonzalezaguilera deeplearningbasedroadsegmentationof3dpointcloudsforassistingroadalignmentparameterization |