Understanding Point Clouds of Electrical Substations Using a Multi-View Multi-Scale Fusion With Transformer

We propose a deep learning-based network model, relying on multi-view multi-scale fusion with transformer, for understanding semantical information of electrical substation point clouds. Our method is inspired by human recognition of 3D objects using information from multiple perspectives and with m...

Full description

Saved in:
Bibliographic Details
Main Authors: Xinghua Wang, Yanxi Meng, Hantuo Dong, Tao Jia
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11028987/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850161315484532736
author Xinghua Wang
Yanxi Meng
Hantuo Dong
Tao Jia
author_facet Xinghua Wang
Yanxi Meng
Hantuo Dong
Tao Jia
author_sort Xinghua Wang
collection DOAJ
description We propose a deep learning-based network model, relying on multi-view multi-scale fusion with transformer, for understanding semantical information of electrical substation point clouds. Our method is inspired by human recognition of 3D objects using information from multiple perspectives and with multiple levels of details, and it is constructed by a dynamic integration of multi-view 3D morphological representation and multi-scale 3D geometric characterization. Additionally, we leverage the channel and spatial attention mechanism to capture the relationship between morphological representations at multiple views and utilize the self and cross attention mechanism to understand the relationship between geometric characterization at multiple scales. To verify our model, experiments were conducted based on our substation point clouds dataset and two benchmark datasets. Our model shows a better performance than the state-of-the-art methods, and it achieves an overall accuracy of 91.15% on our dataset, 93.30% on ModelNet40, and 92.80% on ModelNet40-C, indicating its effectiveness and robustness.
format Article
id doaj-art-1d2c97b8f80c42b48dbbaf7e09ee9a08
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-1d2c97b8f80c42b48dbbaf7e09ee9a082025-08-20T02:22:55ZengIEEEIEEE Access2169-35362025-01-011310049210050310.1109/ACCESS.2025.357822611028987Understanding Point Clouds of Electrical Substations Using a Multi-View Multi-Scale Fusion With TransformerXinghua Wang0Yanxi Meng1Hantuo Dong2Tao Jia3https://orcid.org/0000-0003-4921-6833Department of Infrastructure, Guangdong Power Grid Company Ltd., Guangzhou, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaPlanning Research Center, Guangdong Power Grid Company Ltd., Guangzhou, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, ChinaWe propose a deep learning-based network model, relying on multi-view multi-scale fusion with transformer, for understanding semantical information of electrical substation point clouds. Our method is inspired by human recognition of 3D objects using information from multiple perspectives and with multiple levels of details, and it is constructed by a dynamic integration of multi-view 3D morphological representation and multi-scale 3D geometric characterization. Additionally, we leverage the channel and spatial attention mechanism to capture the relationship between morphological representations at multiple views and utilize the self and cross attention mechanism to understand the relationship between geometric characterization at multiple scales. To verify our model, experiments were conducted based on our substation point clouds dataset and two benchmark datasets. Our model shows a better performance than the state-of-the-art methods, and it achieves an overall accuracy of 91.15% on our dataset, 93.30% on ModelNet40, and 92.80% on ModelNet40-C, indicating its effectiveness and robustness.https://ieeexplore.ieee.org/document/11028987/Deep learningmulti-view multi-scale fusiontransformer3D object classificationelectrical substation point clouds
spellingShingle Xinghua Wang
Yanxi Meng
Hantuo Dong
Tao Jia
Understanding Point Clouds of Electrical Substations Using a Multi-View Multi-Scale Fusion With Transformer
IEEE Access
Deep learning
multi-view multi-scale fusion
transformer
3D object classification
electrical substation point clouds
title Understanding Point Clouds of Electrical Substations Using a Multi-View Multi-Scale Fusion With Transformer
title_full Understanding Point Clouds of Electrical Substations Using a Multi-View Multi-Scale Fusion With Transformer
title_fullStr Understanding Point Clouds of Electrical Substations Using a Multi-View Multi-Scale Fusion With Transformer
title_full_unstemmed Understanding Point Clouds of Electrical Substations Using a Multi-View Multi-Scale Fusion With Transformer
title_short Understanding Point Clouds of Electrical Substations Using a Multi-View Multi-Scale Fusion With Transformer
title_sort understanding point clouds of electrical substations using a multi view multi scale fusion with transformer
topic Deep learning
multi-view multi-scale fusion
transformer
3D object classification
electrical substation point clouds
url https://ieeexplore.ieee.org/document/11028987/
work_keys_str_mv AT xinghuawang understandingpointcloudsofelectricalsubstationsusingamultiviewmultiscalefusionwithtransformer
AT yanximeng understandingpointcloudsofelectricalsubstationsusingamultiviewmultiscalefusionwithtransformer
AT hantuodong understandingpointcloudsofelectricalsubstationsusingamultiviewmultiscalefusionwithtransformer
AT taojia understandingpointcloudsofelectricalsubstationsusingamultiviewmultiscalefusionwithtransformer