DTNLS: 3D Point Cloud Segmentation Based on 2D Image and 3D Point Cloud Double Texture Feature

Panoramic segmentation of 3D point clouds is an essential and challenging technology for robots with 3D detection and measurement capabilities. In order to fuse the color information of 2D image pixels with the spatial position information of the 3D LiDAR point cloud, it is necessary to establish th...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhiguang Liu, Xiaoxiao Yan, Jiahui Zhao, Yong Shi, Fei Yu, Jian Zhao
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10640074/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849426370420015104
author Zhiguang Liu
Xiaoxiao Yan
Jiahui Zhao
Yong Shi
Fei Yu
Jian Zhao
author_facet Zhiguang Liu
Xiaoxiao Yan
Jiahui Zhao
Yong Shi
Fei Yu
Jian Zhao
author_sort Zhiguang Liu
collection DOAJ
description Panoramic segmentation of 3D point clouds is an essential and challenging technology for robots with 3D detection and measurement capabilities. In order to fuse the color information of 2D image pixels with the spatial position information of the 3D LiDAR point cloud, it is necessary to establish the corresponding relationship between the RGB of pixels and the XYZ position of the 3D LiDAR point cloud. We present Double Texture Neighbor LiDAR Segmentation (DTNLS) in this letter. Double texture refers to the color texture of the image and the point cloud texture of the 3D LiDAR. The DTNLS method first uses the color texture feature of the image to segment the pixels and then obtains the clustering center and segmentation boundary outline. 3D LiDAR point cloud texture segmentation takes the clustering center obtained above as the diffusion source, diffuses outwards along the ring LiDAR line, finds the LiDAR point cloud texture boundary features near the image segmentation boundary contour, and finally realizes 3D point cloud segmentation. We carried out quantitative analysis experiments, and the results showed that compared with the best results among the existing mainstream segmentation methods, the proposed DTNLS method improved the accuracy of pedestrian segmentation by 32.2%. The recall improved by 20.12% in pedestrian segmentation. IoU improved by 48.8% for pedestrians. We conducted empirical studies on public datasets and our datasets to demonstrate that DTNLS has broad applicability and better performance in 3D point cloud segmentation than the previous latest techniques, without the need for any 2D images and 3D point cloud training data.
format Article
id doaj-art-1b067081d4d349c2bd9727bc5d5304f3
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-1b067081d4d349c2bd9727bc5d5304f32025-08-20T03:29:27ZengIEEEIEEE Access2169-35362025-01-011310404710405710.1109/ACCESS.2024.344659310640074DTNLS: 3D Point Cloud Segmentation Based on 2D Image and 3D Point Cloud Double Texture FeatureZhiguang Liu0https://orcid.org/0000-0002-9011-5156Xiaoxiao Yan1Jiahui Zhao2Yong Shi3Fei Yu4Jian Zhao5https://orcid.org/0000-0002-5242-4177School of Control and Mechanical Engineering, Tianjin Chengjian University, Tianjin, ChinaCATARC (Tianjin) Automotive Engineering Research Institute Company Ltd., Tianjin, ChinaSchool of Control and Mechanical Engineering, Tianjin Chengjian University, Tianjin, ChinaGuangzhou City Construction College, School of Electrical and Mechanical Engineering, Guangzhou, Guangdong, ChinaSchool of Mechanical Engineering, Hebei University of Technology, Tianjin, ChinaSchool of Control and Mechanical Engineering, Tianjin Chengjian University, Tianjin, ChinaPanoramic segmentation of 3D point clouds is an essential and challenging technology for robots with 3D detection and measurement capabilities. In order to fuse the color information of 2D image pixels with the spatial position information of the 3D LiDAR point cloud, it is necessary to establish the corresponding relationship between the RGB of pixels and the XYZ position of the 3D LiDAR point cloud. We present Double Texture Neighbor LiDAR Segmentation (DTNLS) in this letter. Double texture refers to the color texture of the image and the point cloud texture of the 3D LiDAR. The DTNLS method first uses the color texture feature of the image to segment the pixels and then obtains the clustering center and segmentation boundary outline. 3D LiDAR point cloud texture segmentation takes the clustering center obtained above as the diffusion source, diffuses outwards along the ring LiDAR line, finds the LiDAR point cloud texture boundary features near the image segmentation boundary contour, and finally realizes 3D point cloud segmentation. We carried out quantitative analysis experiments, and the results showed that compared with the best results among the existing mainstream segmentation methods, the proposed DTNLS method improved the accuracy of pedestrian segmentation by 32.2%. The recall improved by 20.12% in pedestrian segmentation. IoU improved by 48.8% for pedestrians. We conducted empirical studies on public datasets and our datasets to demonstrate that DTNLS has broad applicability and better performance in 3D point cloud segmentation than the previous latest techniques, without the need for any 2D images and 3D point cloud training data.https://ieeexplore.ieee.org/document/10640074/Textural featuressegmentation and categorizationdiffusion source3D point cloud segmentation
spellingShingle Zhiguang Liu
Xiaoxiao Yan
Jiahui Zhao
Yong Shi
Fei Yu
Jian Zhao
DTNLS: 3D Point Cloud Segmentation Based on 2D Image and 3D Point Cloud Double Texture Feature
IEEE Access
Textural features
segmentation and categorization
diffusion source
3D point cloud segmentation
title DTNLS: 3D Point Cloud Segmentation Based on 2D Image and 3D Point Cloud Double Texture Feature
title_full DTNLS: 3D Point Cloud Segmentation Based on 2D Image and 3D Point Cloud Double Texture Feature
title_fullStr DTNLS: 3D Point Cloud Segmentation Based on 2D Image and 3D Point Cloud Double Texture Feature
title_full_unstemmed DTNLS: 3D Point Cloud Segmentation Based on 2D Image and 3D Point Cloud Double Texture Feature
title_short DTNLS: 3D Point Cloud Segmentation Based on 2D Image and 3D Point Cloud Double Texture Feature
title_sort dtnls 3d point cloud segmentation based on 2d image and 3d point cloud double texture feature
topic Textural features
segmentation and categorization
diffusion source
3D point cloud segmentation
url https://ieeexplore.ieee.org/document/10640074/
work_keys_str_mv AT zhiguangliu dtnls3dpointcloudsegmentationbasedon2dimageand3dpointclouddoubletexturefeature
AT xiaoxiaoyan dtnls3dpointcloudsegmentationbasedon2dimageand3dpointclouddoubletexturefeature
AT jiahuizhao dtnls3dpointcloudsegmentationbasedon2dimageand3dpointclouddoubletexturefeature
AT yongshi dtnls3dpointcloudsegmentationbasedon2dimageand3dpointclouddoubletexturefeature
AT feiyu dtnls3dpointcloudsegmentationbasedon2dimageand3dpointclouddoubletexturefeature
AT jianzhao dtnls3dpointcloudsegmentationbasedon2dimageand3dpointclouddoubletexturefeature