{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T18:10:33Z","timestamp":1780337433466,"version":"3.54.1"},"reference-count":37,"publisher":"Public Library of Science (PLoS)","issue":"6","license":[{"start":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T00:00:00Z","timestamp":1780272000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"The Natural Science Foundation of Jiangsu Higher Education Institutions of China","award":["25KJB590004"],"award-info":[{"award-number":["25KJB590004"]}]},{"name":"The Jiangsu Province Higher Education Teaching Reform Research Project","award":["2023JSJG624"],"award-info":[{"award-number":["2023JSJG624"]}]},{"name":"The Jiangsu Provincial Education Science Planning Project","award":["ZJCX\/2022\/10"],"award-info":[{"award-number":["ZJCX\/2022\/10"]}]},{"name":"The 333 High-level Talents Training Project of Jiangsu Province","award":["Jiangsu Talent Office No.2"],"award-info":[{"award-number":["Jiangsu Talent Office No.2"]}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>\n                    Object Detection (OD) stands as a fundamental task in the area of autonomous driving environment perception. This study introduces a 3D OD method grounded in deep learning and an improved Euclidean clustering algorithm, aiming to improve the accuracy and efficiency of point cloud segmentation and OD. The core methodological innovations include: (1) the integration of the Cloth Simulation Filter (CSF) for accurate ground and non-ground point separation, combined with a K-Dimensional Tree (KD-Tree) structure and an adaptive parameter mechanism to enhance clustering robustness and efficiency; and (2) an enhanced PointNet architecture incorporating multi-scale grouping (MSG), multi-resolution grouping (MRG), and skip connections to improve local feature extraction and multi-level feature fusion. This method is differentiated from prior works by its holistic integration of density-aware segmentation and hierarchical feature aggregation, addressing key bottlenecks in handling sparse and uneven LiDAR data. The proposed method is rigorously evaluated on the KITTI and NuScenes benchmarks. It achieves segmentation accuracies of 94.96% and 93.12%, with single-frame processing times of 15.63\u2009ms and 17.24\u2009ms, respectively, demonstrating a superior balance of speed and precision compared to traditional Euclidean clustering and other baseline methods. For the 3D OD task, the model attains average detection accuracies of 94.36% and 92.68% on the respective datasets, representing statistically significant improvements (\n                    <jats:italic>p<\/jats:italic>\n                    \u2009&lt;\u20090.001) over the standard PointNet. The detection speed reaches 34 fps and 31 fps, meeting real-time requirements while outperforming existing frameworks in challenging scenarios involving occluded and multi-scale objects. The findings confirm that the proposed framework provides a robust, efficient, and generalizable solution for 3D environmental perception in autonomous driving systems.\n                  <\/jats:p>","DOI":"10.1371\/journal.pone.0348581","type":"journal-article","created":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T17:50:01Z","timestamp":1780336201000},"page":"e0348581","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["3D object detection for vehicle-mounted LiDAR based on deep learning and euclidean clustering algorithm"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-7691-2219","authenticated-orcid":true,"given":"Nan","family":"Zhang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Maolong","family":"Xi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Juan","family":"Fang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fangqin","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,6,1]]},"reference":[{"issue":"3","key":"pone.0348581.ref001","doi-asserted-by":"crossref","first-page":"2152","DOI":"10.1109\/TIV.2024.3444780","article-title":"Precise synchronization between LiDAR and multiple cameras for autonomous driving: an adaptive approach","volume":"10","author":"AK Gurumadaiah","year":"2025","journal-title":"IEEE Trans Intell Veh"},{"issue":"1","key":"pone.0348581.ref002","doi-asserted-by":"crossref","first-page":"962","DOI":"10.1109\/TITS.2022.3167957","article-title":"Lidar point cloud compression, processing and learning for autonomous driving","volume":"24","author":"R Abbasi","year":"2023","journal-title":"IEEE Trans Intell Transport Syst"},{"issue":"2","key":"pone.0348581.ref003","doi-asserted-by":"crossref","first-page":"1699","DOI":"10.1109\/TIV.2022.3213796","article-title":"Performance and challenges of 3D object detection methods in complex scenes for autonomous driving","volume":"8","author":"K Wang","year":"2023","journal-title":"IEEE Trans Intell Veh"},{"issue":"8","key":"pone.0348581.ref004","doi-asserted-by":"crossref","first-page":"2122","DOI":"10.1007\/s11263-023-01784-z","article-title":"Multi-modal 3d object detection in autonomous driving: a survey","volume":"131","author":"Y Wang","year":"2023","journal-title":"Int J Comput Vis"},{"issue":"4","key":"pone.0348581.ref005","doi-asserted-by":"crossref","first-page":"761","DOI":"10.1177\/01423312221093147","article-title":"Survey on deep learning-based 3D object detection in autonomous driving","volume":"45","author":"Z Liang","year":"2022","journal-title":"Transac Institu Measurem Control"},{"issue":"4","key":"pone.0348581.ref006","doi-asserted-by":"crossref","first-page":"3378","DOI":"10.1109\/JSEN.2023.3235830","article-title":"Deep learning-based image 3-D object detection for autonomous driving: review","volume":"23","author":"SY Alaba","year":"2023","journal-title":"IEEE Sensors J"},{"issue":"2","key":"pone.0348581.ref007","doi-asserted-by":"crossref","first-page":"105","DOI":"10.20517\/ir.2021.20","article-title":"Deep learning for LiDAR-only and LiDAR-fusion 3D perception: a survey","volume":"2","author":"D Wu","year":"2022","journal-title":"Intell Robot"},{"issue":"10","key":"pone.0348581.ref008","doi-asserted-by":"crossref","first-page":"18040","DOI":"10.1109\/TITS.2022.3154537","article-title":"CL3D: camera-LiDAR 3D object detection with point feature enhancement and point-guided fusion","volume":"23","author":"C Lin","year":"2022","journal-title":"IEEE Trans Intell Transport Syst"},{"issue":"4","key":"pone.0348581.ref009","doi-asserted-by":"crossref","first-page":"1242","DOI":"10.1109\/TETCI.2023.3259441","article-title":"3D object detection and tracking based on lidar-camera fusion and IMM-UKF algorithm towards highway driving","volume":"7","author":"C Nie","year":"2023","journal-title":"IEEE Trans Emerg Top Comput Intell"},{"issue":"10","key":"pone.0348581.ref010","first-page":"4103","article-title":"Real-time fast channel clustering for LiDAR point cloud","volume":"69","author":"X Zhang","year":"2022","journal-title":"IEEE Trans Circuits Syst II"},{"issue":"2","key":"pone.0348581.ref011","doi-asserted-by":"crossref","first-page":"1141","DOI":"10.1109\/LRA.2023.3236568","article-title":"Mask-based panoptic LiDAR segmentation for autonomous driving","volume":"8","author":"R Marcuzzi","year":"2023","journal-title":"IEEE Robot Autom Lett"},{"issue":"7","key":"pone.0348581.ref012","doi-asserted-by":"crossref","first-page":"10608","DOI":"10.1109\/JSEN.2024.3361509","article-title":"Real-time point cloud clustering algorithm based on roadside LiDAR","volume":"24","author":"J Wu","year":"2024","journal-title":"IEEE Sensors J"},{"issue":"5","key":"pone.0348581.ref013","doi-asserted-by":"crossref","first-page":"1251","DOI":"10.1007\/s12239-024-00112-9","article-title":"The research of 3D point cloud data clustering based on MEMS lidar for autonomous driving","volume":"25","author":"W Yang","year":"2024","journal-title":"Int J Automot Technol"},{"issue":"4","key":"pone.0348581.ref014","doi-asserted-by":"crossref","first-page":"434","DOI":"10.18520\/cs\/v124\/i4\/434-441","article-title":"Three-dimensional point cloud segmentation using a combination of RANSAC and clustering methods","volume":"124","author":"PS Singh","year":"2023","journal-title":"Curr Sci"},{"key":"pone.0348581.ref015","doi-asserted-by":"crossref","first-page":"9317","DOI":"10.1109\/CVPR52729.2023.00899","article-title":"Towards unsupervised object detection from LiDAR point clouds","author":"L Zhang","year":"2023","journal-title":"2023 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)"},{"issue":"3","key":"pone.0348581.ref016","doi-asserted-by":"crossref","first-page":"2893","DOI":"10.1609\/aaai.v36i3.20194","article-title":"Behind the curtain: learning occluded shapes for 3D object detection","volume":"36","author":"Q Xu","year":"2022","journal-title":"AAAI"},{"issue":"2","key":"pone.0348581.ref017","doi-asserted-by":"crossref","first-page":"972","DOI":"10.1609\/aaai.v38i2.27857","article-title":"CMDA: cross-modal and domain adversarial adaptation for LiDAR-based 3D object detection","volume":"38","author":"G Chang","year":"2024","journal-title":"AAAI"},{"issue":"4","key":"pone.0348581.ref018","doi-asserted-by":"crossref","first-page":"2714","DOI":"10.1109\/COMST.2023.3302474","article-title":"A survey on approximate edge AI for energy efficient autonomous driving services","volume":"25","author":"D Katare","year":"2023","journal-title":"IEEE Commun Surv Tutorials"},{"issue":"1","key":"pone.0348581.ref019","doi-asserted-by":"crossref","first-page":"1953","DOI":"10.1109\/TIV.2023.3319985","article-title":"Efficient 3D object detection based on pseudo-LiDAR representation","volume":"9","author":"H Meng","year":"2024","journal-title":"IEEE Trans Intell Veh"},{"issue":"1","key":"pone.0348581.ref020","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3607873","article-title":"Efficient density-peaks clustering algorithms on static and dynamic data in euclidean space","volume":"18","author":"D Amagata","year":"2023","journal-title":"ACM Trans Knowl Discov Data"},{"key":"pone.0348581.ref021","doi-asserted-by":"crossref","first-page":"1","DOI":"10.17509\/gea.v25i1.71587","article-title":"Comparative analysis of DTM extraction from airborne LiDAR point cloud data with adaptive TIN filter, cloth simulation filter, and progressive morphological filter methods","volume":"25","author":"PP Sulung","year":"2025","journal-title":"J Pendidikan Geografi"},{"issue":"5","key":"pone.0348581.ref022","doi-asserted-by":"crossref","first-page":"3587","DOI":"10.1007\/s42835-023-01641-6","article-title":"Power-line extraction and modelling from 3D point clouds data based on KD tree DBSCAN algorithm","volume":"19","author":"YR Pan","year":"2024","journal-title":"J Electr Eng Technol"},{"issue":"5","key":"pone.0348581.ref023","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3723161","article-title":"Low complexity speech secure hash retrieval algorithm based on KDTree nearest neighbor search","volume":"24","author":"Y Huang","year":"2025","journal-title":"ACM Trans Asian Low-Resour Lang Inf Process"},{"issue":"2","key":"pone.0348581.ref024","doi-asserted-by":"crossref","first-page":"175","DOI":"10.18178\/joig.12.2.175-185","article-title":"Improvement of DBSCAN algorithm involving automatic parameters estimation and curvature analysis in 3D point cloud of piled pipe","volume":"12","author":"AR Pratama","year":"2024","journal-title":"J Image Graph"},{"issue":"1","key":"pone.0348581.ref025","doi-asserted-by":"crossref","first-page":"38","DOI":"10.1016\/j.geog.2021.10.002","article-title":"A novel fast classification filtering algorithm for LiDAR point clouds based on small grid density clustering","volume":"13","author":"X Deng","year":"2022","journal-title":"Geod Geodyn"},{"issue":"4","key":"pone.0348581.ref026","doi-asserted-by":"crossref","first-page":"504","DOI":"10.1080\/07038992.2022.2096579","article-title":"Road boundary, curb and surface extraction from 3D Mobile LiDAR point clouds in urban environment","volume":"48","author":"N Wang","year":"2022","journal-title":"Canadian J Remote Sens"},{"issue":"3","key":"pone.0348581.ref027","first-page":"123","article-title":"Advances in real-time object detection and information retrieval: A review","volume":"1","author":"SN Katkade","year":"2023","journal-title":"Intell Artif Intell Appl"},{"issue":"30","key":"pone.0348581.ref028","doi-asserted-by":"crossref","first-page":"611","DOI":"10.1016\/j.ifacol.2025.12.305","article-title":"Autonomous defect detection for point cloud using deep neural network","volume":"59","author":"E Shata","year":"2025","journal-title":"IFAC-PapersOnLine"},{"issue":"5","key":"pone.0348581.ref029","doi-asserted-by":"crossref","first-page":"485","DOI":"10.1108\/SR-01-2022-0022","article-title":"Overview of LiDAR point cloud target detection methods based on deep learning","volume":"42","author":"S Huang","year":"2022","journal-title":"Sens Rev"},{"key":"pone.0348581.ref030","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.cag.2025.104319","article-title":"PointNet with KAN versus PointNet with MLP for 3D classification and segmentation of point sets","volume":"131","author":"A Kashefi","year":"2025","journal-title":"Comput Graph"},{"issue":"7","key":"pone.0348581.ref031","first-page":"9055","article-title":"Sparse tensor-based multiscale representation for point cloud geometry compression","volume":"45","author":"J Wang","year":"2023","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"issue":"1","key":"pone.0348581.ref032","doi-asserted-by":"crossref","DOI":"10.1007\/s12145-024-01543-9","article-title":"DAPNet++: density adaptive PointNet\u2009+\u2009+\u2009for airborne laser scanning data","volume":"18","author":"Z Akbulut","year":"2024","journal-title":"Earth Sci Inform"},{"issue":"370","key":"pone.0348581.ref033","first-page":"1","article-title":"Gradient-free optimization of highly smooth functions: improved analysis and a new algorithm","volume":"25","author":"A Akhavan","year":"2024","journal-title":"J Mach Learn Res"},{"issue":"8","key":"pone.0348581.ref034","doi-asserted-by":"crossref","first-page":"8527","DOI":"10.1109\/TITS.2024.3392783","article-title":"VoPiFNet: voxel-pixel fusion network for multi-class 3D object detection","volume":"25","author":"C-H Wang","year":"2024","journal-title":"IEEE Trans Intell Transport Syst"},{"issue":"3","key":"pone.0348581.ref035","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1007\/s00138-024-01538-y","article-title":"ET-PointPillars: improved PointPillars for 3D object detection based on optimized voxel downsampling","volume":"35","author":"Y Liu","year":"2024","journal-title":"Mach Vis Appl"},{"issue":"3","key":"pone.0348581.ref036","doi-asserted-by":"crossref","first-page":"7286","DOI":"10.1109\/TTE.2023.3346375","article-title":"Voxel RCNN-HA: a point cloud multiobject detection algorithm with hybrid anchors for autonomous driving","volume":"10","author":"H Wang","year":"2023","journal-title":"IEEE T Transp Electrif"},{"issue":"25","key":"pone.0348581.ref037","doi-asserted-by":"crossref","first-page":"35843","DOI":"10.1007\/s11042-021-11660-y","article-title":"E^2-PV-RCNN: improving 3D object detection via enhancing keypoint features","volume":"81","author":"X Li","year":"2022","journal-title":"Multimed Tools Appl"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0348581","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,1]],"date-time":"2026-06-01T17:50:08Z","timestamp":1780336208000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0348581"}},"subtitle":[],"editor":[{"given":"Francesco","family":"Visentin","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,6,1]]},"references-count":37,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,6,1]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0348581","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,1]]}}}