Optimization Method of Airborne LiDAR Individual Tree Segmentation Based on Gaussian Mixture Model

被引:4
|
作者
Zhang, Zhenyu [1 ]
Wang, Jian [1 ]
Li, Zhiyuan [1 ]
Zhao, Youlong [1 ]
Wang, Ruisheng [2 ]
Habib, Ayman [3 ]
机构
[1] Shandong Univ Sci & Technol, Coll Geodesy & Geomatics, Qingdao 266590, Peoples R China
[2] Univ Calgary, Univ Drive NW 2500, Schulich Sch Engn, Dept Geomatics Engn, Calgary, AB T2N 1N4, Canada
[3] Purdue Univ, Lyles Sch Civil Engn, W Lafayette, IN 47907 USA
关键词
airborne LiDAR; Gaussian mixture model; individual tree segmentation; under-segmentation; mean shift; CROWN DELINEATION; POINT CLOUDS; MEAN SHIFT; SINGLE; EFFICIENT; DENSITY; NUMBER;
D O I
10.3390/rs14236167
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Forests are the main part of the terrestrial ecosystem. Airborne LiDAR is fast, comprehensive, penetrating, and contactless and can depict 3D canopy information with a high efficiency and accuracy. Therefore, it plays an important role in forest ecological protection, tree species recognition, carbon sink calculation, etc. Accurate recognition of individual trees in forests is a key step to various application. In real practice, however, the accuracy of individual tree segmentation (ITS) is often compromised by under-segmentation due to the diverse species, obstruction and understory trees typical of a high-density multistoried mixed forest area. Therefore, this paper proposes an ITS optimization method based on Gaussian mixture model for airborne LiDAR data. First, the mean shift (MS) algorithm is used for the initial ITS of the pre-processed airborne LiDAR data. Next, under-segmented samples are extracted by integrated learning, normally segmented samples are classified by morphological approximation, and the approximate distribution uncertainty of the normal samples is described with a covariance matrix. Finally, the class composition among the under-segmented samples is determined, and the under-segmented samples are re-segmented using Gaussian mixture model (GMM) clustering, in light of the optimal covariance matrix of the corresponding categories. Experiments with two datasets, Trento and Qingdao, resulted in ITS recall of 94% and 96%, accuracy of 82% and 91%, and F-scores of 0.87 and 0.93. Compared with the MS algorithm, our method is more accurate and less likely to under-segment individual trees in many cases. It can provide data support for the management and conservation of high-density multistoried mixed forest areas.
引用
收藏
页数:24
相关论文
共 50 条
  • [1] A New Image Segmentation Method for Individual Tree Recognition based on Airborne LiDAR Data
    Tang Feifei
    Liu Xing
    Zhang Yali
    Ruan Zhimin
    [J]. 2010 6TH INTERNATIONAL CONFERENCE ON WIRELESS COMMUNICATIONS NETWORKING AND MOBILE COMPUTING (WICOM), 2010,
  • [2] A topology-based approach to individual tree segmentation from airborne LiDAR data
    Xu, Xin
    Iuricich, Federico
    De Floriani, Leila
    [J]. GEOINFORMATICA, 2023, 27 (04) : 759 - 788
  • [3] A topology-based approach to individual tree segmentation from airborne LiDAR data
    Xin Xu
    Federico Iuricich
    Leila De Floriani
    [J]. GeoInformatica, 2023, 27 : 759 - 788
  • [4] Tree Species Classification Using Airborne LiDAR Data Based on Individual Tree Segmentation and Shape Fitting
    Qian, Chen
    Yao, Chunjing
    Ma, Hongchao
    Xu, Junhao
    Wang, Jie
    [J]. REMOTE SENSING, 2023, 15 (02)
  • [5] An airborne image stabilization method based on the Gaussian mixture model
    Hongbin Deng
    Yunde Jia
    Yihua Xu
    Wei Liang
    [J]. 2007 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN AND CYBERNETICS, VOLS 1-8, 2007, : 2488 - 2492
  • [6] Review on Individual Tree Detection Based on Airborne LiDAR
    Liu Huiling
    Zhang Xiaoli
    Zhang Ying
    Zhu Yunfeng
    Liu Hui
    Wang Longyang
    [J]. LASER & OPTOELECTRONICS PROGRESS, 2018, 55 (08)
  • [7] AN INDIVIDUAL TREE SEGMENTATION ALGORITHM WITHOUT PRIOR-KNOWLEDGE BASED ON AIRBORNE LIDAR DATA
    Li Shihua
    Liu Yuhan
    Zhao Shunda
    Wan Lihong
    [J]. IGARSS 2023 - 2023 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, 2023, : 3225 - 3228
  • [8] Individual Tree Extraction from Terrestrial LiDAR Point Clouds Based on Transfer Learning and Gaussian Mixture Model Separation
    Hui, Zhenyang
    Jin, Shuanggen
    Li, Dajun
    Ziggah, Yao Yevenyo
    Liu, Bo
    [J]. REMOTE SENSING, 2021, 13 (02) : 1 - 32
  • [9] Individual Tree Segmentation Method Based on Mobile Backpack LiDAR Point Clouds
    Comesana-Cebral, Lino
    Martinez-Sanchez, Joaquin
    Lorenzo, Henrique
    Arias, Pedro
    [J]. SENSORS, 2021, 21 (18)
  • [10] The Influence of Vegetation Characteristics on Individual Tree Segmentation Methods with Airborne LiDAR Data
    Yang, Qiuli
    Su, Yanjun
    Jin, Shichao
    Kelly, Maggi
    Hu, Tianyu
    Ma, Qin
    Li, Yumei
    Song, Shilin
    Zhang, Jing
    Xu, Guangcai
    Wei, Jianxin
    Guo, Qinghua
    [J]. REMOTE SENSING, 2019, 11 (23)