Deep-Learning-Based Semantic Segmentation Approach for Point Clouds of Extra-High-Voltage Transmission Lines
The accurate semantic segmentation of point cloud data is the basis for their application in the inspection of extra high-voltage transmission lines (EHVTL). As deep learning evolves, point-wise-based deep neural networks have shown great potential for the semantic segmentation of EHVTL point clouds...
Main Authors: | , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-04-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/15/9/2371 |
_version_ | 1797601724291088384 |
---|---|
author | Hao Yu Zhengyang Wang Qingjie Zhou Yuxuan Ma Zhuo Wang Huan Liu Chunqing Ran Shengli Wang Xinghua Zhou Xiaobo Zhang |
author_facet | Hao Yu Zhengyang Wang Qingjie Zhou Yuxuan Ma Zhuo Wang Huan Liu Chunqing Ran Shengli Wang Xinghua Zhou Xiaobo Zhang |
author_sort | Hao Yu |
collection | DOAJ |
description | The accurate semantic segmentation of point cloud data is the basis for their application in the inspection of extra high-voltage transmission lines (EHVTL). As deep learning evolves, point-wise-based deep neural networks have shown great potential for the semantic segmentation of EHVTL point clouds. However, EHVTL point cloud data are characterized by a large data volume and significant class imbalance. Therefore, the down-sampling method and point cloud feature extraction method used in current point-wise-based deep neural networks hardly meet the needs of computational accuracy and efficiency. In this paper, we proposed a two-step down-sampling method and a point cloud feature extraction method based on local feature aggregation of the point clouds after down-sampling in each layer of the model (LFAPAD). We then established a deep neural network named PowerLine-Net for the semantic segmentation of the EHVTL point clouds. Furthermore, in order to test and analyze the performance of PowerLine-Net, we constructed a point cloud dataset for the EHVTL scenes. Using this dataset and the Semantic3D dataset, we implemented network parameter testing, semantic segmentation, and an accuracy comparison of different networks based on PowerLine-Net. The results illustrate that the semantic segmentation model proposed in this paper has a high computational efficiency and accuracy in the semantic segmentation of EHVTL point clouds. Compared with conventional deep neural networks, including PointCNN, KPConv, SPG, PointNet++, and RandLA-Net, PowerLine-Net also achieves a higher accuracy in the semantic segmentation of EHVTL point clouds. Moreover, based on the results predicted by PowerLine-Net, the risk point detection for EHVTL point clouds has been achieved, which demonstrates the important value of this network in practical applications. In addition, as shown by the results of Semantic3D, PowerLine-Net also achieves a high segmentation accuracy, which proves its powerful capability and wide applicability in semantic segmentation for the point clouds of large-scale scenes. |
first_indexed | 2024-03-11T04:07:43Z |
format | Article |
id | doaj.art-316def11ac9b483aaee94ffe0531ea1f |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-11T04:07:43Z |
publishDate | 2023-04-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-316def11ac9b483aaee94ffe0531ea1f2023-11-17T23:39:13ZengMDPI AGRemote Sensing2072-42922023-04-01159237110.3390/rs15092371Deep-Learning-Based Semantic Segmentation Approach for Point Clouds of Extra-High-Voltage Transmission LinesHao Yu0Zhengyang Wang1Qingjie Zhou2Yuxuan Ma3Zhuo Wang4Huan Liu5Chunqing Ran6Shengli Wang7Xinghua Zhou8Xiaobo Zhang9College of Ocean Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, ChinaCollege of Ocean Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, ChinaFirst Institute of Oceanography, Ministry of Natural Resources, Qingdao 266061, ChinaCollege of Ocean Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, ChinaCollege of Ocean Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, ChinaCollege of Ocean Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, ChinaCollege of Ocean Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, ChinaCollege of Ocean Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, ChinaFirst Institute of Oceanography, Ministry of Natural Resources, Qingdao 266061, ChinaCollege of Ocean Science and Engineering, Shandong University of Science and Technology, Qingdao 266590, ChinaThe accurate semantic segmentation of point cloud data is the basis for their application in the inspection of extra high-voltage transmission lines (EHVTL). As deep learning evolves, point-wise-based deep neural networks have shown great potential for the semantic segmentation of EHVTL point clouds. However, EHVTL point cloud data are characterized by a large data volume and significant class imbalance. Therefore, the down-sampling method and point cloud feature extraction method used in current point-wise-based deep neural networks hardly meet the needs of computational accuracy and efficiency. In this paper, we proposed a two-step down-sampling method and a point cloud feature extraction method based on local feature aggregation of the point clouds after down-sampling in each layer of the model (LFAPAD). We then established a deep neural network named PowerLine-Net for the semantic segmentation of the EHVTL point clouds. Furthermore, in order to test and analyze the performance of PowerLine-Net, we constructed a point cloud dataset for the EHVTL scenes. Using this dataset and the Semantic3D dataset, we implemented network parameter testing, semantic segmentation, and an accuracy comparison of different networks based on PowerLine-Net. The results illustrate that the semantic segmentation model proposed in this paper has a high computational efficiency and accuracy in the semantic segmentation of EHVTL point clouds. Compared with conventional deep neural networks, including PointCNN, KPConv, SPG, PointNet++, and RandLA-Net, PowerLine-Net also achieves a higher accuracy in the semantic segmentation of EHVTL point clouds. Moreover, based on the results predicted by PowerLine-Net, the risk point detection for EHVTL point clouds has been achieved, which demonstrates the important value of this network in practical applications. In addition, as shown by the results of Semantic3D, PowerLine-Net also achieves a high segmentation accuracy, which proves its powerful capability and wide applicability in semantic segmentation for the point clouds of large-scale scenes.https://www.mdpi.com/2072-4292/15/9/2371extra-high-voltage transmission linespoint cloud semantic segmentationdeep learningtwo-step down-samplingpoint cloud feature extraction |
spellingShingle | Hao Yu Zhengyang Wang Qingjie Zhou Yuxuan Ma Zhuo Wang Huan Liu Chunqing Ran Shengli Wang Xinghua Zhou Xiaobo Zhang Deep-Learning-Based Semantic Segmentation Approach for Point Clouds of Extra-High-Voltage Transmission Lines Remote Sensing extra-high-voltage transmission lines point cloud semantic segmentation deep learning two-step down-sampling point cloud feature extraction |
title | Deep-Learning-Based Semantic Segmentation Approach for Point Clouds of Extra-High-Voltage Transmission Lines |
title_full | Deep-Learning-Based Semantic Segmentation Approach for Point Clouds of Extra-High-Voltage Transmission Lines |
title_fullStr | Deep-Learning-Based Semantic Segmentation Approach for Point Clouds of Extra-High-Voltage Transmission Lines |
title_full_unstemmed | Deep-Learning-Based Semantic Segmentation Approach for Point Clouds of Extra-High-Voltage Transmission Lines |
title_short | Deep-Learning-Based Semantic Segmentation Approach for Point Clouds of Extra-High-Voltage Transmission Lines |
title_sort | deep learning based semantic segmentation approach for point clouds of extra high voltage transmission lines |
topic | extra-high-voltage transmission lines point cloud semantic segmentation deep learning two-step down-sampling point cloud feature extraction |
url | https://www.mdpi.com/2072-4292/15/9/2371 |
work_keys_str_mv | AT haoyu deeplearningbasedsemanticsegmentationapproachforpointcloudsofextrahighvoltagetransmissionlines AT zhengyangwang deeplearningbasedsemanticsegmentationapproachforpointcloudsofextrahighvoltagetransmissionlines AT qingjiezhou deeplearningbasedsemanticsegmentationapproachforpointcloudsofextrahighvoltagetransmissionlines AT yuxuanma deeplearningbasedsemanticsegmentationapproachforpointcloudsofextrahighvoltagetransmissionlines AT zhuowang deeplearningbasedsemanticsegmentationapproachforpointcloudsofextrahighvoltagetransmissionlines AT huanliu deeplearningbasedsemanticsegmentationapproachforpointcloudsofextrahighvoltagetransmissionlines AT chunqingran deeplearningbasedsemanticsegmentationapproachforpointcloudsofextrahighvoltagetransmissionlines AT shengliwang deeplearningbasedsemanticsegmentationapproachforpointcloudsofextrahighvoltagetransmissionlines AT xinghuazhou deeplearningbasedsemanticsegmentationapproachforpointcloudsofextrahighvoltagetransmissionlines AT xiaobozhang deeplearningbasedsemanticsegmentationapproachforpointcloudsofextrahighvoltagetransmissionlines |