点云切片结合聚类算法的TLS单木探测方法研究

doi:10.12302/j.issn.1000-2006.202206035

国家林草科技领军期刊
中国精品科技期刊
中国高校百佳科技期刊
江苏省新闻出版政府奖期刊奖
RCCSE林学权威期刊（A+）
CSCD核心期刊
Scopus数据库收录期刊
中文核心期刊
SCD核心期刊

作者加群：102861116

微信公众号：南京林业大学学报

高级检索

南京林业大学学报（自然科学版） ›› 2024, Vol. 48 ›› Issue (4): 113-122.doi: 10.12302/j.issn.1000-2006.202206035

所属专题：专题报道Ⅲ：智慧林业之森林可视化研究

• 专题报道Ⅲ：智慧林业之森林可视化研究（执行主编李凤日、张怀清、曹林） • 上一篇下一篇

点云切片结合聚类算法的TLS单木探测方法研究

易静¹(), 马开森¹^,², 向建平³, 唐杰¹, 蒋馥根¹, 陈松¹, 孙华¹^,^*()

1.中南林业科技大学林业遥感信息工程研究中心,林业遥感大数据与生态安全湖南省重点实验室,南方森林资源经营与监测国家林业和草原局重点实验室,湖南长沙 410004
2.湖南科技大学地理空间信息技术国家地方联合工程实验室,湖南湘潭 411201
3.中南林业科技大学芦头实验林场,湖南岳阳 414000

收稿日期:2022-06-20 修回日期:2022-08-29 出版日期:2024-07-30 发布日期:2024-08-05
通讯作者: *孙华(sunhua@csuft.edu.cn),教授。
作者简介:
易静(yijing@csuft.edu.cn)。
基金资助:
国家自然科学基金面上项目(31971578);湖南省自然科学基金面上项目(2022JJ30078)

Research on TLS single tree detection method based on point cloud slicing combined with clustering algorithm

YI Jing¹(), MA Kaisen¹^,², XIANG Jianping³, TANG Jie¹, JIANG Fugen¹, CHEN Song¹, SUN Hua¹^,^*()

1. Research Center of Forestry Remote Sensing & Information Engineering, Central South University of Forestry and Technology, Key Laboratory of Forestry Remote Sensing Based Big Data & Ecological Security for Hunan Province, Key Laboratory of National Forestry and Grassland Administration on Forest Resources Management and Monitoring in Southern China, Changsha 410004,China
2. National-Local Joint Engineering Laboratory of Geo-Spatial Information Technology, Hunan University of Science & Technology, Xiangtan 411201, China
3. Lutou Experimental Forest Farm, Central South University of Forestry and Technology, Yueyang 414000, China

Received:2022-06-20 Revised:2022-08-29 Online:2024-07-30 Published:2024-08-05

摘要/Abstract

摘要：

【目的】针对直接使用地面激光雷达生成的冠层高度模型(canopy height model,CHM)和归一化点云(normalized point cloud,NPC)在复杂林分探测中存在单木探测能力不足的问题,研究引入点云切片结合聚类的方法以提高单木探测精度。【方法】以广西壮族自治区6个不同林分密度的人工林样地为研究对象,利用地面激光扫描获取样地的归一化点云数据,提取高度在1.3 m处的点云切片,分别采用基于密度噪声应用空间聚类(DBSCAN)和均值漂移聚类(MS)算法对切片中的树干点云进行聚类。利用野外实测调查数据进行精度验证,并与基于CHM的局部最大值算法和基于NPC的点云分割算法(point cloud segmentation,PCS)的探测结果对比,评价和分析不同探测方法的适用性与参数敏感性。【结果】所有方法均可获得良好的探测结果,各样地的最优总体探测精度得分F≥ 0.86;点云切片结合聚类算法的单木探测方法结果最优。DBSCAN算法的聚类阈值(Eps)和均值漂移算法的聚类半径可显著影响单木探测率,最大Eps取决于最大林木间距,聚类半径接近最大单木胸径时的探测结果最优。【结论】基于点云切片结合聚类算法的单木探测能提高下层林木探测率,可有效改善高密度林分的单木探测精度,为不同林分的单木探测方法选择提供参考。

关键词: 地面激光扫描, 单木探测, 点云切片, 聚类算法, 林木参数提取, 人工林

Abstract:

【Objective】To solve the problem that a canopy height model (CHM) and normalized point cloud (NPC) directly generated by terrestrial laser scanning (TLS) are not capable of detecting individual trees in complex stands, this study introduced the method of point cloud slicing combined with clustering to improve the detection accuracy.【Method】In this study, six sample plots in a plantation with different stand densities in Guangxi Zhuang Autonomous Region, China, were used as the research objects. First, the NPC data of a sample plot obtained by TLS were used to extract point cloud slices at a height of 1.3 m, and then the density-based spatial clustering of applications with noise (DBSCAN) and mean shift algorithms were used to cluster the tree trunk point clouds in the slices. The accuracy was verified by the field survey data, and the detection results were compared with those of the local maximum algorithm based on a CHM, and a point cloud segmentation algorithm based on an NPC. The applicability and parameter sensitivity of the different detection methods were evaluated and analyzed.【Result】Satisfactory detection results were obtained by all methods, and the optimal detection accuracy F-score was ≥ 0.86 for each sample plot. The individual tree detection method using point cloud slicing combined with a clustering algorithm produced better results. The clustering threshold epsilon neighborhood (Eps) value of the DBSCAN algorithm and the clustering radius r of the mean shift algorithm significantly affected the individual tree detection rate, with the maximum Eps depending on the maximum stand spacing and optimum results when r was close to the maximum individual tree diameter at breast height.【Conclusion】Individual tree detection based on point cloud slicing combined with a clustering algorithm can increase the detection rate of understory trees lower forest, effectively improve the accuracy of single tree detection in dense stands, and provide a reference for the selection of single tree detection methods in different forest stands.

Key words: terrestrial laser scanning, single tree detection, point cloud slicing, clustering algorithm, extraction of forest parameters, plantation

中图分类号:

S758
TP391

易静,马开森,向建平,等. 点云切片结合聚类算法的TLS单木探测方法研究[J]. 南京林业大学学报（自然科学版）, 2024, 48(4): 113-122.

YI Jing, MA Kaisen, XIANG Jianping, TANG Jie, JIANG Fugen, CHEN Song, SUN Hua. Research on TLS single tree detection method based on point cloud slicing combined with clustering algorithm[J].Journal of Nanjing Forestry University (Natural Science Edition), 2024, 48(4): 113-122.DOI: 10.12302/j.issn.1000-2006.202206035.

图/表 9

表1

图1

表2

图2

图3

图4

图5

图6

图7

参考文献 37

[1]	曹林, 佘光辉, 代劲松, 等. 激光雷达技术估测森林生物量的研究现状及展望[J]. 南京林业大学学报(自然科学版), 2013, 37(3):163-169.
	CAO L, SHE G H, DAI J S, et al. Status and prospects of the LiDAR-based forest biomass estimation[J]. J Nanjing For Univ (Nat Sci Ed), 2013, 37(3):163-169.DOI: 10.3969/j.issn.1000-2006.2013.03.029.
[2]	李增元, 刘清旺, 庞勇. 激光雷达森林参数反演研究进展[J]. 遥感学报, 2016, 20(5):1138-1150.
	LI Z Y, LIU Q W, PANG Y. Review on forest parameters inversion using LiDAR[J]. J Remote Sens, 2016, 20(5):1138-1150.DOI: 10.11834/jrs.20165130.
[3]	吴楠, 李增元, 廖声熙, 等. 国内外林业遥感应用研究概况与展望[J]. 世界林业研究, 2017, 30(6):34-40.
	WU N, LI Z Y, LIAO S X, et al. Current situation and prospect of research on application of remote sensing to forestry[J]. World For Res, 2017, 30(6):34-40.DOI: 10.13348/j.cnki.sjlyyj.2017.0075.y.
[4]	刘鲁霞, 庞勇. 机载激光雷达和地基激光雷达林业应用现状[J]. 世界林业研究, 2014, 27(1):49-56.
	LIU L X, PANG Y. Applications of airborne laser scanning and terrestrial laser scanning to forestry[J]. World For Res, 2014, 27(1):49-56.DOI: 10.13348/j.cnki.sjlyyj.2014.01.009.
[5]	花伟成, 田佳榕, 孙心雨, 等. 基于TLS数据的杨树削度方程建立及材积估算[J]. 南京林业大学学报(自然科学版), 2021, 45(4):41-48.
	HUA W C, TIAN J R, SUN X Y, et al. Assessing the stem taper function and volume estimation of poplar (Populus)by terrestrial laser scanning[J]. J Nanjing For Univ (Nat Sci Ed), 2021, 45(4):41-48.DOI: 10.12302/j.issn.1000-2006.202006023.
[6]	蒋佳文, 温小荣, 顾海波, 等. 基于多站扫描的点云特征参数与材积结构动态分析[J]. 南京林业大学学报(自然科学版), 2019, 43(6):83-90.
	JIANG J W, WEN X R, GU H B, et al. Dynamic analysis of point cloud characteristic parameters and volume structure based on multi-station scan[J]. J Nanjing For Univ (Nat Sci Ed), 2019, 43(6):83-90.DOI: 10.3969/j.issn.1000-2006.201901020.
[7]	XU D D, WANG H B, XU W X, et al. LiDAR applications to estimate forest biomass at individual tree scale:opportunities,challenges and future perspectives[J]. Forests, 2021, 12(5):550.DOI: 10.3390/f12050550.
[8]	刘会玲, 张晓丽, 张莹, 等. 机载激光雷达单木识别研究进展[J]. 激光与光电子学进展, 2018, 55(8):40-48.
	LIU H L, ZHANG X L, ZHANG Y, et al. Review on individual tree detection based on airborne LiDAR[J]. Laser Optoelectron Prog, 2018, 55(8):40-48.DOI: 10.3788/LOP55.082805.
[9]	WANG L, GONG P, BIGING G S. Individual tree-crown delineation and treetop detection in high-spatial-resolution aerial imagery[J]. Photogramm Eng Remote Sensing, 2004, 70(3):351-357.DOI: 10.14358/pers.70.3.351.
[10]	刘方舟, 刘浩, 云挺. 基于分水岭优化思想的单木信息分割算法[J]. 林业工程学报, 2020, 5(5):109-116.
	LIU F Z, LIU H, YUN T. Individual tree crown separation using the improved watershed method[J]. J For Eng, 2020, 5(5):109-116.DOI: 10.13360/j.issn.2096-1359.202001021.
[11]	郭庆华, 刘瑾, 陶胜利, 等. 激光雷达在森林生态系统监测模拟中的应用现状与展望[J]. 科学通报, 2014, 59(6):459-478.
	GUO Q H, LIU J, TAO S L, et al. Perspectives and prospects of LiDAR in forest ecosystem monitoring and modeling[J]. Chin Sci Bull, 2014, 59(6):459-478.DOI: 10.1360/972013-592.
[12]	LI W K, GUO Q H, JAKUBOWSKI M K, et al. A new method for segmenting individual trees from the lidar point cloud[J]. Photogramm Eng Remote Sensing, 2012, 78(1):75-84.DOI: 10.14358/pers.78.1.75.
[13]	LIANG X L, KANKARE V, HYYPPÄ J, et al. Terrestrial laser scanning in forest inventories[J]. ISPRS J Photogramm Remote Sens, 2016, 115:63-77.DOI: 10.1016/j.isprsjprs.2016.01.006.
[14]	LU X C, GUO Q H, LI W K, et al. A bottom-up approach to segment individual deciduous trees using leaf-off lidar point cloud data[J]. ISPRS J Photogramm Remote Sens, 2014, 94:1-12.DOI: 10.1016/j.isprsjprs.2014.03.014.
[15]	CHEN W, HU X B, CHEN W, et al. Airborne LiDAR remote sensing for individual tree forest inventory using trunk detection-aided mean shift clustering techniques[J]. Remote Sens, 2018, 10(7):1078.DOI: 10.3390/rs10071078.
[16]	MADHULATHA T S. An overview on clustering methods[J]. IOSR J Eng, 2012, 2(4):719-725.DOI: 10.9790/3021-0204719725.
[17]	林秀云, 孙圆, 刘晨曦, 等. 依据地面激光扫描数据的杉木材积建模与造材[J]. 东北林业大学学报, 2022, 50(1):33-39.
	LIN X Y, SUN Y, LIU C X, et al. Volume modeling and merchantable volume for Cunninghamia lanceolata using terrestrial laser scanner[J]. J Northeast For Univ, 2022, 50(1):33-39.DOI: 10.13759/j.cnki.dlxb.2022.01.012.
[18]	WANG C X, JI M, WANG J, et al. An improved DBSCAN method for LiDAR data segmentation with automatic eps estimation[J]. Sensors, 2019, 19(1):172.DOI: 10.3390/s19010172.
[19]	麻卫峰, 王金亮, 麻源源, 等. 改进K均值聚类的点云林木胸径提取[J]. 测绘科学, 2021, 46(9):122-129.
	MA W F, WANG J L, MA Y Y, et al. An improved K-means clustering method for DBH extraction from point cloud[J]. Science of Surveying and Mapping, 2021, 46(9):122-129.DOI: 10.16251/j.cnki.1009-2307.2021.09.016.
[20]	TAO S L, WU F F, GUO Q H, et al. Segmenting tree crowns from terrestrial and mobile LiDAR data by exploring ecological theories[J]. ISPRS J Photogramm Remote Sens, 2015, 110:66-76.DOI: 10.1016/j.isprsjprs.2015.10.007.
[21]	DUONG T, BECK G, AZZAG H, et al. Nearest neighbour estimators of density derivatives,with application to mean shift clustering[J]. Pattern Recognit Lett, 2016, 80:224-230.DOI: 10.1016/j.patrec.2016.06.021.
[22]	朱德海. 点云库PCL学习教程[M]. 北京: 北京航空航天大学出版社, 2012:189-191.
	ZHU D H. PCL learning course of point cloud library[M]. Beijing: Beijing University of Aeronautics & Astronautics Press, 2012:189-191.
[23]	ZHAO X Q, GUO Q H, SU Y J, et al. Improved progressive TIN densification filtering algorithm for airborne LiDAR data in forested areas[J]. ISPRS J Photogramm Remote Sens, 2016, 117:79-91.DOI: 10.1016/j.isprsjprs.2016.03.016.
[24]	KHOSRAVIPOUR A, SKIDMORE A K, ISENBURG M. Generating spike-free digital surface models using LiDAR raw point clouds:a new approach for forestry applications[J]. Int J Appl Earth Obs Geoinf, 2016, 52:104-114.DOI: 10.1016/j.jag.2016.06.005.
[25]	WULDER M, NIEMANN K O, GOODENOUGH D G. Local maximum filtering for the extraction of tree locations and basal area from high spatial resolution imagery[J]. Remote Sens Environ, 2000, 73(1):103-114.DOI: 10.1016/S0034-4257(00)00101-2.
[26]	POPESCU S C, WYNNE R H, NELSON R F. Estimating plot-level tree heights with lidar:local filtering with a canopy-height based variable window size[J]. Comput Electron Agric, 2002, 37(1/2/3):71-95.DOI: 10.1016/S0168-1699(02)00121-7.
[27]	CHEN Q, BALDOCCHI D, GONG P, et al. Isolating individual trees in a savanna woodland using small footprint lidar data[J]. Photogramm Eng Remote Sensing, 2006, 72(8):923-932.DOI: 10.14358/pers.72.8.923.
[28]	TANG J, JIANG F G, LONG Y, et al. Identification of the yield of Camellia oleifera based on color space by the optimized mean shift clustering algorithm using terrestrial laser scanning[J]. Remote Sens, 2022, 14(3):642.DOI: 10.3390/rs14030642.
[29]	WATT P J, DONOGHUE D N M. Measuring forest structure with terrestrial laser scanning[J]. Int J Remote Sens, 2005, 26(7):1437-1446.DOI: 10.1080/01431160512331337961.
[30]	MA K S, CHEN Z X, FU L Y, et al. Performance and sensitivity of individual tree segmentation methods for UAV-LiDAR in multiple forest types[J]. Remote Sens, 2022, 14(2):298.DOI: 10.3390/rs14020298.
[31]	李响, 甄贞, 赵颖慧. 基于局域最大值法单木位置探测的适宜模型研究[J]. 北京林业大学学报, 2015, 37(3):27-33.
	LI X, ZHEN Z, ZHAO Y H. Suitable model of detecting the position of individual treetop based on local maximum method[J]. J Beijing For Univ, 2015, 37(3):27-33.DOI: 10.13332/j.1000-1522.20140313.
[32]	LISIEWICZ M, KAMINSKA A, STERENCZAK K. Recognition of specified errors of individual tree detection methods based on canopy height model[J]. Remote Sens Appl, 2022, 25:100690.DOI: 10.1016/j.rsase.2021.100690.
[33]	ZHEN Z, QUACKENBUSH L, ZHANG L J. Trends in automatic individual tree crown detection and delineation: evolution of LiDAR data[J]. Remote Sens, 2016, 8(4):333.DOI: 10.3390/rs8040333.
[34]	LIU J B, LIANG X L, HYYPPÄ J, et al. Automated matching of multiple terrestrial laser scans for stem mapping without the use of artificial references[J]. Int J Appl Earth Obs Geoinf, 2017, 56:13-23.DOI: 10.1016/j.jag.2016.11.003.
[35]	WANG Y S, HYYPPÄ J, LIANG X L, et al. International benchmarking of the individual tree detection methods for modeling 3-D canopy structure for silviculture and forest ecology using airborne laser scanning[J]. IEEE Trans Geosci Remote Sens, 2016, 54(9):5011-5027.DOI: 10.1109/TGRS.2016.2543225.
[36]	YANG Q L, SU Y J, JIN S C, et al. The influence of vegetation characteristics on individual tree segmentation methods with airborne LiDAR data[J]. Remote Sens, 2019, 11(23):2880.DOI: 10.3390/rs11232880.
[37]	杨海城. 基于ULS和TLS的天然次生林不同林层单木参数估测及对比[D]. 哈尔滨: 东北林业大学, 2021.
	YANG H C. Estimation and comparison of individual tree parameters for different forest canopies in natural secondary forest based on ULS and TLS[D]. Harbin: Northeast Forestry University, 2021.

样地号 sample plot No.	优势树种 dominant tree species	林木株树 number of trees	密度/ (株·hm^-2) density	林木分布 distribution of forest trees	胸径/cm DBH
样地号 sample plot No.	优势树种 dominant tree species	林木株树 number of trees	密度/ (株·hm^-2) density	林木分布 distribution of forest trees	平均值 mean	最小值 min	最大值 max	平均值 mean	最小值 min	最大值 max
1	杉木	28	420	离散	20.2	9.4	28.5	12.5	6.3	17.6
2	桉树	52	780	聚集	9.0	5.5	11.5	11.4	8.2	13.5
3	桉树	75	1 125	均匀	8.2	5.0	11.0	11.1	6.5	13.8
4	杉木	94	1 410	均匀	10.1	5.3	16.5	7.6	5.2	11.5
5	桉树	103	1 545	局部聚集	11.2	5.6	20.9	15.8	4.1	19.5
6	杉木	114	1 710	局部聚集	11.4	5.1	40.0	8.8	3.7	18.5

样地号 sample plot number	方法 method	探测株数 number of detected	正确探测株数 number of correctively detected	召回率/% recall	准确率/% precision	总体精度得分F F-score
1	LM	28	27	96.43	96.43	0.96
	PCS	27	27	96.43	100.00	0.98
	DBSCAN	28	28	100.00	100.00	1.00
	MS	28	28	100.00	100.00	1.00
2	LM	46	37	71.15	80.43	0.76
	PCS	47	38	73.08	80.85	0.77
	DBSCAN	52	47	90.38	90.38	0.90
	MS	56	48	92.31	85.71	0.89
3	LM	75	61	81.33	81.33	0.81
	PCS	71	60	80.00	84.51	0.82
	DBSCAN	75	68	90.67	90.67	0.91
	MS	75	68	90.67	90.67	0.91
4	LM	101	78	82.98	77.23	0.80
	PCS	110	84	89.36	76.36	0.82
	DBSCAN	111	92	97.87	82.88	0.90
	MS	96	86	91.49	89.58	0.91
5	LM	84	73	70.87	86.90	0.78
	PCS	79	71	68.93	89.87	0.78
	DBSCAN	104	90	87.38	86.54	0.87
	MS	107	92	89.32	85.98	0.88
6	LM	112	89	78.07	79.46	0.79
	PCS	114	91	79.82	79.82	0.80
	DBSCAN	108	94	82.46	87.04	0.85
	MS	116	99	86.84	85.34	0.86

点云切片结合聚类算法的TLS单木探测方法研究

Research on TLS single tree detection method based on point cloud slicing combined with clustering algorithm

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 37

相关文章 15

编辑推荐

Metrics

本文评价

作者

读者

审稿

[1]	王梓萌, 阮宏华, 吴小巧, 杨艳, 谢友超, 沈彩芹, 丁学农, 曹国华. 模拟氮沉降对杨树人工林土壤跳虫群落的影响[J]. 南京林业大学学报（自然科学版）, 2024, 48(4): 243-253.
[2]	赵金满, 韩馨悦, 程瑞明, 张志东. 塞罕坝自然保护区华北落叶松和樟子松人工林健康评价[J]. 南京林业大学学报（自然科学版）, 2024, 48(3): 199-206.
[3]	武燕, 黄青, 刘讯, 郑睿, 岑佳宝, 丁波, 张运林, 符裕红. 西南喀斯特地区马尾松人工林林龄对土壤理化性质的影响[J]. 南京林业大学学报（自然科学版）, 2024, 48(3): 99-107.
[4]	鲁旭东, 董禹然, 李垚, 毛岭峰. 中国亚热带杉木人工林不同林分发育阶段的群落构建机制[J]. 南京林业大学学报（自然科学版）, 2024, 48(1): 67-73.
[5]	颜铮明, 阮宏华, 廖家辉, 石珂, 倪娟平, 曹国华, 沈彩芹, 丁学农, 赵小龙, 庄鑫. 不同林龄杨树人工林地表甲虫群落多样性特征[J]. 南京林业大学学报（自然科学版）, 2023, 47(6): 236-242.
[6]	王宇, 易艳灵, 刘海, 文晓晨, 李天一, 尹海锋, 李贤伟, 范川. 两种采伐方式对马尾松人工林林分空间结构的影响[J]. 南京林业大学学报（自然科学版）, 2023, 47(5): 138-146.
[7]	李双娴, 陆鑫, 多杰才仁, 张怀清, 薛联凤, 云挺. 一种新的地面激光点云中树木叶面积计算方法[J]. 南京林业大学学报（自然科学版）, 2023, 47(5): 28-38.
[8]	沈浩, 姜姜, 周晨, 潘庆全. 江西石城不同起源阔叶林碳储量驱动因子分析[J]. 南京林业大学学报（自然科学版）, 2023, 47(4): 185-190.
[9]	孙宇, 李凤日, 谢龙飞, 董利虎. 基于林分及地形因子的落叶松人工林林分生物量模型构建[J]. 南京林业大学学报（自然科学版）, 2023, 47(3): 129-136.
[10]	储安婷, 宁卓, 杨红强. 林业碳汇对人工林最优轮伐期的影响--以杉木和落叶松为例[J]. 南京林业大学学报（自然科学版）, 2023, 47(3): 225-233.
[11]	宋磊, 金星姬, PUKKALA Timo, 李凤日. 长白落叶松人工林多目标经营模式研究[J]. 南京林业大学学报（自然科学版）, 2023, 47(2): 150-158.
[12]	杨永超, 段文标, 陈立新, 曲美学, 王亚飞, 王美娟, 石金永, 潘磊. 模拟氮磷沉降和凋落物处理对两种林型红松林土壤有机碳组分的影响[J]. 南京林业大学学报（自然科学版）, 2023, 47(1): 57-66.
[13]	徐放, 杨晓慧, 潘文, 廖焕琴, 杨会肖, 张卫华, 朱报著, 陈新宇, 徐斌, 王裕霞. 红锥树干节疤分布及修枝对生长的影响[J]. 南京林业大学学报（自然科学版）, 2022, 46(5): 121-126.
[14]	李惠芝, 关庆伟, 赵家豪, 李俊杰, 王磊, 李凤凤, 左兴平, 陈斌. 地形对麻栎人工林土壤肥力质量的影响[J]. 南京林业大学学报（自然科学版）, 2022, 46(5): 161-168.
[15]	赵凯歌, 周正虎, 金鹰, 王传宽. 长期氮添加对落叶松和水曲柳人工林土壤碳、氮、磷含量和胞外酶活性的影响[J]. 南京林业大学学报（自然科学版）, 2022, 46(5): 177-184.