A Supervoxel-Based Random Forest Method for Robust and Effective Airborne LiDAR Point Cloud Classification

被引:14
|
作者
Liao, Lingfeng [1 ]
Tang, Shengjun [1 ]
Liao, Jianghai [1 ]
Li, Xiaoming [1 ]
Wang, Weixi [1 ]
Li, Yaxin [2 ]
Guo, Renzhong [1 ]
机构
[1] Shenzhen Univ, Sch Architecture & Urban Planning, Key Lab Urban Land Resources Monitoring & Simulat, Res Inst Smart Cities,Minist Nat Resources, Shenzhen 518060, Peoples R China
[2] Hong Kong Polytech Univ, Shenzhen Res Inst, Shenzhen 518057, Peoples R China
基金
中国国家自然科学基金;
关键词
point cloud classification; supervoxel; random forest; feature fusion; segmentation; OBJECT DETECTION; IMAGERY;
D O I
10.3390/rs14061516
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
As an essential part of point cloud processing, autonomous classification is conventionally used in various multifaceted scenes and non-regular point distributions. State-of-the-art point cloud classification methods mostly process raw point clouds, using a single point as the basic unit and calculating point cloud features by searching local neighbors via the k-neighborhood method. Such methods tend to be computationally inefficient and have difficulty obtaining accurate feature descriptions due to inappropriate neighborhood selection. In this paper, we propose a robust and effective point cloud classification approach that integrates point cloud supervoxels and their locally convex connected patches into a random forest classifier, which effectively improves the point cloud feature calculation accuracy and reduces the computational cost. Considering the different types of point cloud feature descriptions, we divide features into three categories (point-based, eigen-based, and grid-based) and accordingly design three distinct feature calculation strategies to improve feature reliability. Two International Society of Photogrammetry and Remote Sensing benchmark tests show that the proposed method achieves state-of-the-art performance, with average F1-scores of 89.16 and 83.58, respectively. The successful classification of point clouds with great variation in elevation also demonstrates the reliability of the proposed method in challenging scenes.
引用
收藏
页数:18
相关论文
共 50 条
  • [1] Fisher Vector Encoding of Supervoxel-Based Features for Airborne LiDAR Data Classification
    Akwensi, Perpetual Hope
    Kang, Zhizhong
    Yang, Juntao
    [J]. IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2020, 17 (03) : 504 - 508
  • [2] ROBUST AND EFFECTIVE AIRBORNE LIDAR POINT CLOUD CLASSIFICATION BASED ON HYBRID FEATURES
    Liao, L. F.
    Tang, S. J.
    Liao, J. H.
    Wang, W. X.
    Li, X. M.
    Guo, R. Z.
    [J]. XXIV ISPRS CONGRESS IMAGING TODAY, FORESEEING TOMORROW, COMMISSION II, 2022, 43-B2 : 229 - 235
  • [3] Individual tree segmentation for airborne LiDAR point cloud data using spectral clustering and supervoxel-based algorithm
    Wang W.
    Pang Y.
    Du L.
    Zhang Z.
    Liang X.
    [J]. National Remote Sensing Bulletin, 2022, 26 (08) : 1650 - 1661
  • [4] Supervoxel-based and Cost-Effective Active Learning for Point Cloud Semantic Segmentation
    Ye, Shanding
    Fu, Yongjian
    Lin, Hu
    Yin, Zhe
    Pan, Zhijie
    [J]. 2022 IEEE 25TH INTERNATIONAL CONFERENCE ON INTELLIGENT TRANSPORTATION SYSTEMS (ITSC), 2022, : 1030 - 1036
  • [5] Airborne LiDAR Point Cloud Filtering Algorithm Based on Supervoxel Ground Saliency
    Fan, Weiwei
    Liu, Xinyi
    Zhang, Yongjun
    Yue, Dongdong
    Wang, Senyuan
    Zhong, Jiachen
    [J]. ISPRS ANNALS OF THE PHOTOGRAMMETRY, REMOTE SENSING AND SPATIAL INFORMATION SCIENCES: VOLUME X-2-2024, 2024, : 73 - 79
  • [6] Supervoxel-Based Region Growing Segmentation for Point Cloud Data
    Luo, Nan
    Jiang, Yuanyuan
    Wang, Quan
    [J]. INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE, 2021, 35 (03)
  • [7] Classification of LiDAR Point Clouds Using Supervoxel-Based Detrended Feature and Perception-Weighted Graphical Model
    Xu, Yusheng
    Ye, Zhen
    Yao, Wei
    Huang, Rong
    Tong, Xiaohua
    Hoegner, Ludwig
    Stilla, Uwe
    [J]. IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2020, 13 : 72 - 88
  • [8] A method of classification for airborne full waveform LiDAR data based on random forest
    [J]. Zhang, A. (zhangaw98@163.com), 1600, Science Press (40):
  • [9] Supervoxel-based extraction and classification of pole-like objects from MLS point cloud data
    Li, Jintao
    Cheng, Xiaojun
    [J]. OPTICS AND LASER TECHNOLOGY, 2022, 146
  • [10] Airborne LiDAR Point Cloud Classification Based on Transfer Learning
    Zhao, Chuan
    Yu, Donghang
    Xu, Junfeng
    Zhang, Baoming
    Li, Daoji
    [J]. ELEVENTH INTERNATIONAL CONFERENCE ON DIGITAL IMAGE PROCESSING (ICDIP 2019), 2019, 11179