PointNet++ Network Architecture with Individual Point Level and Global Features on Centroid for ALS Point Cloud Classification
Airborne laser scanning (ALS) point cloud has been widely used in the fields of ground powerline surveying, forest monitoring, urban modeling, and so on because of the great convenience it brings to people’s daily life. However, the sparsity and uneven distribution of point clouds increases the diff...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-01-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/3/472 |
| _version_ | 1827596980903215104 |
|---|---|
| author | Yang Chen Guanlan Liu Yaming Xu Pai Pan Yin Xing |
| author_facet | Yang Chen Guanlan Liu Yaming Xu Pai Pan Yin Xing |
| author_sort | Yang Chen |
| collection | DOAJ |
| description | Airborne laser scanning (ALS) point cloud has been widely used in the fields of ground powerline surveying, forest monitoring, urban modeling, and so on because of the great convenience it brings to people’s daily life. However, the sparsity and uneven distribution of point clouds increases the difficulty of setting uniform parameters for semantic classification. The PointNet++ network is an end-to-end learning network for irregular point data and highly robust to small perturbations of input points along with corruption. It eliminates the need to calculate costly handcrafted features and provides a new paradigm for 3D understanding. However, each local region in the output is abstracted by its centroid and local feature that encodes the centroid’s neighborhood. The feature learned on the centroid point may not contain relevant information of itself for random sampling, especially in large-scale neighborhood balls. Moreover, the centroid point’s global-level information in each sample layer is also not marked. Therefore, this study proposed a modified PointNet++ network architecture which concentrates the point-level and global features on the centroid point towards the local features to facilitate classification. The proposed approach also utilizes a modified Focal Loss function to solve the extremely uneven category distribution on ALS point clouds. An elevation- and distance-based interpolation method is also proposed for the objects in ALS point clouds which exhibit discrepancies in elevation distributions. The experiments on the Vaihingen dataset of the International Society for Photogrammetry and Remote Sensing and the GML(B) 3D dataset demonstrate that the proposed method which provides additional contextual information to support classification achieves high accuracy with simple discriminative models and new state-of-the-art performance in power line categories. |
| first_indexed | 2024-03-09T03:20:41Z |
| format | Article |
| id | doaj.art-ad26edbf6a8f463084e9391edfb4b52c |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T03:20:41Z |
| publishDate | 2021-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-ad26edbf6a8f463084e9391edfb4b52c2023-12-03T15:11:39ZengMDPI AGRemote Sensing2072-42922021-01-0113347210.3390/rs13030472PointNet++ Network Architecture with Individual Point Level and Global Features on Centroid for ALS Point Cloud ClassificationYang Chen0Guanlan Liu1Yaming Xu2Pai Pan3Yin Xing4School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, ChinaSchool of Geodesy and Geomatics, Wuhan University, Wuhan 430079, ChinaSchool of Geodesy and Geomatics, Wuhan University, Wuhan 430079, ChinaSchool of Geodesy and Geomatics, Wuhan University, Wuhan 430079, ChinaSchool of Earth Sciences and Engineering, Hohai University, Nanjing 211100, ChinaAirborne laser scanning (ALS) point cloud has been widely used in the fields of ground powerline surveying, forest monitoring, urban modeling, and so on because of the great convenience it brings to people’s daily life. However, the sparsity and uneven distribution of point clouds increases the difficulty of setting uniform parameters for semantic classification. The PointNet++ network is an end-to-end learning network for irregular point data and highly robust to small perturbations of input points along with corruption. It eliminates the need to calculate costly handcrafted features and provides a new paradigm for 3D understanding. However, each local region in the output is abstracted by its centroid and local feature that encodes the centroid’s neighborhood. The feature learned on the centroid point may not contain relevant information of itself for random sampling, especially in large-scale neighborhood balls. Moreover, the centroid point’s global-level information in each sample layer is also not marked. Therefore, this study proposed a modified PointNet++ network architecture which concentrates the point-level and global features on the centroid point towards the local features to facilitate classification. The proposed approach also utilizes a modified Focal Loss function to solve the extremely uneven category distribution on ALS point clouds. An elevation- and distance-based interpolation method is also proposed for the objects in ALS point clouds which exhibit discrepancies in elevation distributions. The experiments on the Vaihingen dataset of the International Society for Photogrammetry and Remote Sensing and the GML(B) 3D dataset demonstrate that the proposed method which provides additional contextual information to support classification achieves high accuracy with simple discriminative models and new state-of-the-art performance in power line categories.https://www.mdpi.com/2072-4292/13/3/472PointNet++ networkpoint-level and global information on centroid pointmodified focal loss functionelevation- and distance-based interpolationALS point clouds |
| spellingShingle | Yang Chen Guanlan Liu Yaming Xu Pai Pan Yin Xing PointNet++ Network Architecture with Individual Point Level and Global Features on Centroid for ALS Point Cloud Classification Remote Sensing PointNet++ network point-level and global information on centroid point modified focal loss function elevation- and distance-based interpolation ALS point clouds |
| title | PointNet++ Network Architecture with Individual Point Level and Global Features on Centroid for ALS Point Cloud Classification |
| title_full | PointNet++ Network Architecture with Individual Point Level and Global Features on Centroid for ALS Point Cloud Classification |
| title_fullStr | PointNet++ Network Architecture with Individual Point Level and Global Features on Centroid for ALS Point Cloud Classification |
| title_full_unstemmed | PointNet++ Network Architecture with Individual Point Level and Global Features on Centroid for ALS Point Cloud Classification |
| title_short | PointNet++ Network Architecture with Individual Point Level and Global Features on Centroid for ALS Point Cloud Classification |
| title_sort | pointnet network architecture with individual point level and global features on centroid for als point cloud classification |
| topic | PointNet++ network point-level and global information on centroid point modified focal loss function elevation- and distance-based interpolation ALS point clouds |
| url | https://www.mdpi.com/2072-4292/13/3/472 |
| work_keys_str_mv | AT yangchen pointnetnetworkarchitecturewithindividualpointlevelandglobalfeaturesoncentroidforalspointcloudclassification AT guanlanliu pointnetnetworkarchitecturewithindividualpointlevelandglobalfeaturesoncentroidforalspointcloudclassification AT yamingxu pointnetnetworkarchitecturewithindividualpointlevelandglobalfeaturesoncentroidforalspointcloudclassification AT paipan pointnetnetworkarchitecturewithindividualpointlevelandglobalfeaturesoncentroidforalspointcloudclassification AT yinxing pointnetnetworkarchitecturewithindividualpointlevelandglobalfeaturesoncentroidforalspointcloudclassification |