PEMCNet: An Efficient Multi-Scale Point Feature Fusion Network for 3D LiDAR Point Cloud Classification

Point cloud classification plays a significant role in Light Detection and Ranging (LiDAR) applications. However, most available multi-scale feature learning networks for large-scale 3D LiDAR point cloud classification tasks are time-consuming. In this paper, an efficient deep neural architecture de...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Genping Zhao, Weiguang Zhang, Yeping Peng, Heng Wu, Zhuowei Wang, Lianglun Cheng
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	LiDAR point cloud classification deep learning Science Q
Acceso en línea:	https://doaj.org/article/1e0a845ab8e6467bb45e5028053e439e
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:1e0a845ab8e6467bb45e5028053e439e
record_format	dspace
spelling	oai:doaj.org-article:1e0a845ab8e6467bb45e5028053e439e2021-11-11T18:53:41ZPEMCNet: An Efficient Multi-Scale Point Feature Fusion Network for 3D LiDAR Point Cloud Classification10.3390/rs132143122072-4292https://doaj.org/article/1e0a845ab8e6467bb45e5028053e439e2021-10-01T00:00:00Zhttps://www.mdpi.com/2072-4292/13/21/4312https://doaj.org/toc/2072-4292Point cloud classification plays a significant role in Light Detection and Ranging (LiDAR) applications. However, most available multi-scale feature learning networks for large-scale 3D LiDAR point cloud classification tasks are time-consuming. In this paper, an efficient deep neural architecture denoted as Point Expanded Multi-scale Convolutional Network (PEMCNet) is developed to accurately classify the 3D LiDAR point cloud. Different from traditional networks for point cloud processing, PEMCNet includes successive Point Expanded Grouping (PEG) units and Absolute and Relative Spatial Embedding (ARSE) units for representative point feature learning. The PEG unit enables us to progressively increase the receptive field for each observed point and aggregate the feature of a point cloud at different scales but without increasing computation. The ARSE unit following the PEG unit furthermore realizes representative encoding of points relationship, which effectively preserves the geometric details between points. We evaluate our method on both public datasets (the Urban Semantic 3D (US3D) dataset and Semantic3D benchmark dataset) and our new collected Unmanned Aerial Vehicle (UAV) based LiDAR point cloud data of the campus of Guangdong University of Technology. In comparison with four available state-of-the-art methods, our methods ranked first place regarding both efficiency and accuracy. It was observed on the public datasets that with a 2% increase in classification accuracy, over 26% improvement of efficiency was achieved at the same time compared to the second efficient method. Its potential value is also tested on the newly collected point cloud data with over 91% of classification accuracy and 154 ms of processing time.Genping ZhaoWeiguang ZhangYeping PengHeng WuZhuowei WangLianglun ChengMDPI AGarticleLiDARpoint cloudclassificationdeep learningScienceQENRemote Sensing, Vol 13, Iss 4312, p 4312 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	LiDAR point cloud classification deep learning Science Q
spellingShingle	LiDAR point cloud classification deep learning Science Q Genping Zhao Weiguang Zhang Yeping Peng Heng Wu Zhuowei Wang Lianglun Cheng PEMCNet: An Efficient Multi-Scale Point Feature Fusion Network for 3D LiDAR Point Cloud Classification
description	Point cloud classification plays a significant role in Light Detection and Ranging (LiDAR) applications. However, most available multi-scale feature learning networks for large-scale 3D LiDAR point cloud classification tasks are time-consuming. In this paper, an efficient deep neural architecture denoted as Point Expanded Multi-scale Convolutional Network (PEMCNet) is developed to accurately classify the 3D LiDAR point cloud. Different from traditional networks for point cloud processing, PEMCNet includes successive Point Expanded Grouping (PEG) units and Absolute and Relative Spatial Embedding (ARSE) units for representative point feature learning. The PEG unit enables us to progressively increase the receptive field for each observed point and aggregate the feature of a point cloud at different scales but without increasing computation. The ARSE unit following the PEG unit furthermore realizes representative encoding of points relationship, which effectively preserves the geometric details between points. We evaluate our method on both public datasets (the Urban Semantic 3D (US3D) dataset and Semantic3D benchmark dataset) and our new collected Unmanned Aerial Vehicle (UAV) based LiDAR point cloud data of the campus of Guangdong University of Technology. In comparison with four available state-of-the-art methods, our methods ranked first place regarding both efficiency and accuracy. It was observed on the public datasets that with a 2% increase in classification accuracy, over 26% improvement of efficiency was achieved at the same time compared to the second efficient method. Its potential value is also tested on the newly collected point cloud data with over 91% of classification accuracy and 154 ms of processing time.
format	article
author	Genping Zhao Weiguang Zhang Yeping Peng Heng Wu Zhuowei Wang Lianglun Cheng
author_facet	Genping Zhao Weiguang Zhang Yeping Peng Heng Wu Zhuowei Wang Lianglun Cheng
author_sort	Genping Zhao
title	PEMCNet: An Efficient Multi-Scale Point Feature Fusion Network for 3D LiDAR Point Cloud Classification
title_short	PEMCNet: An Efficient Multi-Scale Point Feature Fusion Network for 3D LiDAR Point Cloud Classification
title_full	PEMCNet: An Efficient Multi-Scale Point Feature Fusion Network for 3D LiDAR Point Cloud Classification
title_fullStr	PEMCNet: An Efficient Multi-Scale Point Feature Fusion Network for 3D LiDAR Point Cloud Classification
title_full_unstemmed	PEMCNet: An Efficient Multi-Scale Point Feature Fusion Network for 3D LiDAR Point Cloud Classification
title_sort	pemcnet: an efficient multi-scale point feature fusion network for 3d lidar point cloud classification
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/1e0a845ab8e6467bb45e5028053e439e
work_keys_str_mv	AT genpingzhao pemcnetanefficientmultiscalepointfeaturefusionnetworkfor3dlidarpointcloudclassification AT weiguangzhang pemcnetanefficientmultiscalepointfeaturefusionnetworkfor3dlidarpointcloudclassification AT yepingpeng pemcnetanefficientmultiscalepointfeaturefusionnetworkfor3dlidarpointcloudclassification AT hengwu pemcnetanefficientmultiscalepointfeaturefusionnetworkfor3dlidarpointcloudclassification AT zhuoweiwang pemcnetanefficientmultiscalepointfeaturefusionnetworkfor3dlidarpointcloudclassification AT liangluncheng pemcnetanefficientmultiscalepointfeaturefusionnetworkfor3dlidarpointcloudclassification
_version_	1718431690764845056

PEMCNet: An Efficient Multi-Scale Point Feature Fusion Network for 3D LiDAR Point Cloud Classification

Ejemplares similares