Scattered Train Bolt Point Cloud Segmentation Based on Hierarchical Multi-Scale Feature Learning
In view of the difficulty of using raw 3D point clouds for component detection in the railway field, this paper designs a point cloud segmentation model based on deep learning together with a point cloud preprocessing mechanism. First, a special preprocessing algorithm is designed to resolve the pro...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/4/2019 |
| _version_ | 1797618286873018368 |
|---|---|
| author | Ni Zeng Jinlong Li Yu Zhang Xiaorong Gao Lin Luo |
| author_facet | Ni Zeng Jinlong Li Yu Zhang Xiaorong Gao Lin Luo |
| author_sort | Ni Zeng |
| collection | DOAJ |
| description | In view of the difficulty of using raw 3D point clouds for component detection in the railway field, this paper designs a point cloud segmentation model based on deep learning together with a point cloud preprocessing mechanism. First, a special preprocessing algorithm is designed to resolve the problems of noise points, acquisition errors, and large data volume in the actual point cloud model of the bolt. The algorithm uses the point cloud adaptive weighted guided filtering for noise smoothing according to the noise characteristics. Then retaining the key points of the point cloud, this algorithm uses the octree to partition the point cloud and carries out iterative farthest point sampling in each partition for obtaining the standard point cloud model. The standard point cloud model is then subjected to hierarchical multi-scale feature extraction to obtain global features, which are combined with local features through a self-attention mechanism, while linear interpolation is used to further expand the perceptual field of local features of the model as a basis for segmentation, and finally the segmentation is completed. Experiments show that the proposed algorithm could deal with the scattered bolt point cloud well, realize the segmentation of train bolt and background, and could achieve high segmentation accuracy, which has important practical significance for train safety detection. |
| first_indexed | 2024-03-11T08:10:56Z |
| format | Article |
| id | doaj.art-7ea428fa69d04695b70bc4e18f1dc5a3 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T08:10:56Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-7ea428fa69d04695b70bc4e18f1dc5a32023-11-16T23:09:13ZengMDPI AGSensors1424-82202023-02-01234201910.3390/s23042019Scattered Train Bolt Point Cloud Segmentation Based on Hierarchical Multi-Scale Feature LearningNi Zeng0Jinlong Li1Yu Zhang2Xiaorong Gao3Lin Luo4School of Physical Science and Technology, Southwest Jiaotong University, Chengdu 610031, ChinaSchool of Physical Science and Technology, Southwest Jiaotong University, Chengdu 610031, ChinaSchool of Physical Science and Technology, Southwest Jiaotong University, Chengdu 610031, ChinaSchool of Physical Science and Technology, Southwest Jiaotong University, Chengdu 610031, ChinaSchool of Physical Science and Technology, Southwest Jiaotong University, Chengdu 610031, ChinaIn view of the difficulty of using raw 3D point clouds for component detection in the railway field, this paper designs a point cloud segmentation model based on deep learning together with a point cloud preprocessing mechanism. First, a special preprocessing algorithm is designed to resolve the problems of noise points, acquisition errors, and large data volume in the actual point cloud model of the bolt. The algorithm uses the point cloud adaptive weighted guided filtering for noise smoothing according to the noise characteristics. Then retaining the key points of the point cloud, this algorithm uses the octree to partition the point cloud and carries out iterative farthest point sampling in each partition for obtaining the standard point cloud model. The standard point cloud model is then subjected to hierarchical multi-scale feature extraction to obtain global features, which are combined with local features through a self-attention mechanism, while linear interpolation is used to further expand the perceptual field of local features of the model as a basis for segmentation, and finally the segmentation is completed. Experiments show that the proposed algorithm could deal with the scattered bolt point cloud well, realize the segmentation of train bolt and background, and could achieve high segmentation accuracy, which has important practical significance for train safety detection.https://www.mdpi.com/1424-8220/23/4/2019point clouddeep learningbolt segmentationdenosingdownsampling |
| spellingShingle | Ni Zeng Jinlong Li Yu Zhang Xiaorong Gao Lin Luo Scattered Train Bolt Point Cloud Segmentation Based on Hierarchical Multi-Scale Feature Learning Sensors point cloud deep learning bolt segmentation denosing downsampling |
| title | Scattered Train Bolt Point Cloud Segmentation Based on Hierarchical Multi-Scale Feature Learning |
| title_full | Scattered Train Bolt Point Cloud Segmentation Based on Hierarchical Multi-Scale Feature Learning |
| title_fullStr | Scattered Train Bolt Point Cloud Segmentation Based on Hierarchical Multi-Scale Feature Learning |
| title_full_unstemmed | Scattered Train Bolt Point Cloud Segmentation Based on Hierarchical Multi-Scale Feature Learning |
| title_short | Scattered Train Bolt Point Cloud Segmentation Based on Hierarchical Multi-Scale Feature Learning |
| title_sort | scattered train bolt point cloud segmentation based on hierarchical multi scale feature learning |
| topic | point cloud deep learning bolt segmentation denosing downsampling |
| url | https://www.mdpi.com/1424-8220/23/4/2019 |
| work_keys_str_mv | AT nizeng scatteredtrainboltpointcloudsegmentationbasedonhierarchicalmultiscalefeaturelearning AT jinlongli scatteredtrainboltpointcloudsegmentationbasedonhierarchicalmultiscalefeaturelearning AT yuzhang scatteredtrainboltpointcloudsegmentationbasedonhierarchicalmultiscalefeaturelearning AT xiaoronggao scatteredtrainboltpointcloudsegmentationbasedonhierarchicalmultiscalefeaturelearning AT linluo scatteredtrainboltpointcloudsegmentationbasedonhierarchicalmultiscalefeaturelearning |