基于机载LiDAR数据的崇礼冬奥核心区单木分割算法适宜性分析

doi:10.11707/j.1001-7488.20221012

摘要/Abstract

摘要：

目的: 探究3种基于机载LiDAR数据的单木分割算法获取冬奥核心区新造林油松和中龄林华北落叶松单木信息的适宜性, 分析3种算法的单木分割精度和单木树高信息估算精度, 探索冬奥核心区最优单木分割方法, 为掌握冬奥核心区森林结构信息和制定森林经营管理措施提供技术支持。方法: 利用冬奥核心区典型样地机载LiDAR数据, 应用基于点云的距离判别聚类法、分水岭分割算法和双正切角树冠识别算法, 结合地面调查数据、正射影像与人工目视解译, 采用树冠探测率(r)、准确率(p)和总体精度(F)3个评价指标分析算法的单木分割精度; 单木配准后, 分析外业实测树高与机载LiDAR估算树高之间的相关性, 综合评估3种算法获取冬奥核心区单木信息的适宜性。结果: 1) 3种单木分割算法在新造林油松中总体精度很高(F=0.90~0.93), 在中龄林华北落叶松中总体精度较高(F=0.72~0.75); 2) 对于新造林油松, 基于点云的距离判别聚类法总体精度最高(F=0.93), 优于分水岭分割算法(F=0.90)和双正切角树冠识别算法(F=0.90); 对于中龄林华北落叶松, 双正切角树冠识别算法总体精度最高(F=0.75), 优于基于点云的距离判别聚类法(F=0.72)和分水岭分割算法(F=0.70); 3) 外业实测树高与机载LiDAR估算树高之间的相关性分析表明, 基于点云的距离判别聚类法在新造林油松中提取的单木树高相关性最优, 双正切角树冠识别算法在中龄林华北落叶松中提取的单木树高相关性最优。结论: 通过对冬奥核心区不同类型林地的机载LiDAR数据进行单木分割, 体现出基于点云的距离判别聚类法、分水岭分割算法和双正切角树冠识别算法的适宜性, 在新造林油松中, 基于点云的距离判别聚类法适宜性表现最优; 在中龄林华北落叶松中, 双正切角树冠识别算法适宜性表现最优。

关键词: 激光LiDAR, 单木分割, 冬奥, 分水岭分割算法, 树冠识别

Abstract:

Objective: The suitability of three tree segmentation algorithms based on airborne LiDAR data in obtaining information of Pinus tabulaeformis and middle-aged Larix principis-rupprechtii in the core area of the Winter Olympic Games was studied, and the accuracy of individual tree segmentation and tree height extraction of the three methods were analyzed, with the aims to explore the optimal single tree segmentation method in the core area of the Winter Olympic Games and also to provide technical supports for mastering the forest structure information and formulating forest management measures in the core area of the Winter Olympic Games. Method: Based on the airborne LiDAR data of the sample plots in the core area of the Winter Olympic Games, the point cloud-based cluster segmentation algorithm, watershed segmentation algorithm and double-tangent crown recognition algorithm were applied, combined with sample plot survey data, orthophoto image and artificial visual interpretation, and tree crown detection rate(r), accuracy(p) and overall accuracy(F) were used to analyze the single tree segmentation accuracy of the algorithms; After individual tree registration, the correlation between field measured tree height and LiDAR estimated tree height was analyzed, and the suitability of three tree segmentation algorithms in the core area of Winter Olympic Games was comprehensively evaluated. Result: 1) The overall segmentation accuracy of the three tree segmentation methods was very high in newly afforestation Pinus tabulaeformis(F=0.90-0.93) and higher in middle-aged forest Larix principis-rupprechtii(F=0.72-0.75). 2) For newly afforestation Pinus tabulaeformis, the segmentation accuracy of point cloud-based cluster segmentation algorithm was the highest(F=0.93), which was better than that of watershed segmentation algorithm(F=0.90) and double-tangent crown recognition algorithm(F=0.90). For middle-aged Larix principis-rupprechtii, the segmentation accuracy of double-tangent crown recognition algorithm was the highest(F=0.75), which was better than that of the point cloud-based cluster segmentation algorithm(F=0.72) and watershed segmentation algorithm(F=0.70). 3) The correlation analysis between field measured tree height and airborne LiDAR estimated tree height showed that the correlation of single tree height extracted from newly afforestation Pinus tabulaeformis by point cloud-based cluster segmentation algorithm was the best, and the correlation of single tree height extracted by double-tangent crown recognition algorithm was the best in middle-aged Larix principis-rupprechtii. Conclusion: Through the single tree segmentation of the airborne LiDAR of different types of forest land in the core area of the Winter Olympic Games, the suitability of the point cloud-based cluster segmentation, watershed segmentation algorithm and double-tangent crown recognition algorithm could be reflected. The best segmentation was achieved using the point cloud-based cluster segmentation algorithm for newly planted Pinus tabulaeformis, and the best suitability segmentation was achieved using the double-tangent crown recognition algorithm for middle-aged Larix principis-rupprechtii.

Key words: laser LiDAR, single tree segmentation, Winter Olympic Games, watershed segmentation algorithm, crown recognition

中图分类号:

S757

谢栋博,刘清旺,雷雅凯,于航,杨旭平,符利勇. 基于机载LiDAR数据的崇礼冬奥核心区单木分割算法适宜性分析[J]. 林业科学, 2022, 58(10): 121-130.

Dongbo Xie,Qingwang Liu,Yakai Lei,Hang Yu,Xuping Yang,Liyong Fu. Suitability Analysis of Single Tree Segmentation Algorithm in the Core Area of Winter Olympic Games Based on Airborne LiDAR Data[J]. Scientia Silvae Sinicae, 2022, 58(10): 121-130.

图/表 12

图1

图2

表1

图3

图4

图5

图6

表2

图7

表3

图8

表4

参考文献 0

	胡凯龙, 刘清旺, 庞勇, 等. 基于机载激光雷达校正的ICESat/GLAS数据森林冠层高度估测. 农业工程学报, 2017, 33 (16): 88- 95.
	Hu K L , Liu Q W , Pang Y , et al. Forest canopy height estimation based on ICESat/GLAS data by airborne LiDAR. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33 (16): 88- 95.
	霍朗宁, 张晓丽. 基于机载LiDAR点云多层聚类的单木信息提取及其精度评价. 林业科学, 2021, 57 (1): 85- 94.
	Huo L N , Zhang X L . Individual tree information extraction and accuracy evaluation based on airborne LiDAR point cloud by multilayer clustering method. Scientia Silvae Sinicae, 2021, 57 (1): 85- 94.
	李平昊, 申鑫, 代劲松, 等. 机载激光雷达人工林单木分割方法比较和精度分析. 林业科学, 2018, 54 (12): 127- 136.
	Li P H , Shen X , Dai J S , et al. Comparisons and accuracy assessments of LiDAR-based tree segmentation approaches in planted forests. Scientia Silvae Sinicae, 2018, 54 (12): 127- 136.
	刘国华, 傅伯杰, 方精云. 中国森林碳动态及其对全球碳平衡的贡献. 生态学报, 2000, 20 (5): 733- 740.
	Liu G H , Fu B J , Fang J Y . Carbon dynamics of Chinese forests and its contribution to global carbon balance. Acta Ecologica Sinica, 2000, 20 (5): 733- 740.
	刘清旺, 李世明, 李增元, 等. 无人机激光雷达与摄影测量林业应用研究进展. 林业科学, 2017, 53 (7): 134- 148.
	Liu Q W , Li S M , Li Z Y , et al. Review on the applications of UAV-based LiDAR and photogrammetry in forestry. Scientia Silvae Sinicae, 2017, 53 (7): 134- 148.
	刘清旺, 李增元, 陈尔学, 等. 利用机载激光雷达数据提取单株木树高和树冠. 北京林业大学学报, 2008, 30 (6): 83- 89.
	Liu Q W , Li Z Y , Chen E X , et al. Extracting height and crown of individual tree using airborne LiDAR data. Journal of Beijing Forestry University, 2008, 30 (6): 83- 89.
	申鑫, 曹林, 佘光辉. 高光谱与高空间分辨率遥感数据的亚热带森林生物量反演. 遥感学报, 2016, 20 (6): 1446- 1460.
	Shen X , Cao L , She G H . Subtropical forest biomass estimation based on hyperspectral and high-resolution remotely sensed data. Journal of Remote Sensing, 2016, 20 (6): 1446- 1460.
	吴项乾, 曹林, 申鑫, 等. 基于无人机激光雷达的银杏人工林有效叶面积指数估测. 林业科学, 2020, 56 (1): 74- 86.
	Wu X Q , Cao L , Shen X , et al. Estimation of effective leaf area index using UAV-based LiDAR in Ginkgo plantations. Scientia Silvae Sinicae, 2020, 56 (1): 74- 86.
	王瑞瑞, 李怡燃, 石伟, 等. 基于LiDAR数据的单木提取算法研究. 西北林学院学报, 2021, 36 (3): 182- 189.
	Wang R R , Li Y R , Shi W , et al. Single wood extraction algorithm based on LiDAR data. Journal of Northwest Forestry University, 2021, 36 (3): 182- 189.
	王轶夫, 岳天祥, 赵明伟, 等. 机载LiDAR数据的树高识别算法与应用分析. 地球信息科学学报, 2014, 16 (6): 958- 963.
	Wang Y F , Yue T X , Zhao M W , et al. Study of factors impacting the tree height extraction based on airborne LiDAR data. Journal of Geo-Information Science, 2014, 16 (6): 958- 963.
	杨秋丽. 2018. 机载LiDAR单木分割方法对比及其影响因素分析. 乌鲁木齐: 新疆大学.
	Yang Q L. 2018. Comparison of airborne LiDAR individual tree segmentation methods and analysis of influencing factors. Urumqi: Xinjiang University. [in Chinese]
	赵峰, 庞勇, 李增元, 等. 机载激光雷达和航空数码影像单木树高提取. 林业科学, 2009, 45 (10): 81- 87.
	Zhao F , Pang Y , Li Z Y , et al. Extraction of individual tree height using a combination of aerial digital camera imagery and LiDAR. Scientia Silvae Sinicae, 2009, 45 (10): 81- 87.
	Alexander C . Delineating tree crowns from airborne laser scanning point cloud data using Delaunay triangulation. International Journal of Remote Sensing, 2009, 30 (14): 3843- 3848.
	Ayrey E , Fraver S , Kershaw J A , et al. Layer stacking: a novel algorithm for individual forest tree segmentation from LiDAR point clouds. Canadian Journal of Remote Sensing, 2017, 43 (1): 16- 27.
	Chen Q , Baldocchi D , Gong P , et al. Isolating individual trees in a savanna woodland using small footprint LiDAR data. Photogrammetric Engineering & Remote Sensing, 2006, 72 (8): 923- 932.
	Duncanson L I , Cook B D , Hurtt G C , et al. An efficient, multi-layered crown delineation algorithm for mapping individual tree structure across multiple ecosystems. Remote Sensing of Environment, 2014, 154, 378- 386.
	Eysn J , Hollaus M , Lindberg E , et al. A benchmark of LiDAR-based single tree detection methods using heterogeneous forest data from the alpine space. Forests, 2015, 6 (12): 1721- 1747.
	Guo Q , Li W , Yu H . Effects of topographic variability and LiDAR sampling density on several DEM interpolation methods. Photogrammetric Engineering & Remote Sensing, 2010, 76 (6): 701- 712.
	Guo Q H , Yan J , Zhao X Q , et al. Improved progressive TIN densification filtering algorithm for airborne LiDAR data in forested areas. ISPRS Journal of Photogrammetry and Remote Sensing, 2016, 117, 79- 91.
	Hosoi F , Matsugami H , Watanuki K , et al. Accurate detection of tree apexes in coniferous canopies from airborne scanning light detection and ranging images based on crown-extraction filtering. Journal of Applied Remote Sensing, 2012, 6 (1): 063502-1- 063502-13.
	Jennings S , Brown N , Sheil D . Assessing forest canopies and understorey illumination: canopy closure, canopy cover and other measures. Forestry: An International Journal of Forest Research, 1999, 72 (1): 59- 74.
	Kaartinen H , Hyyppä J , Yu X W , et al. An international comparison of individual tree detection and extraction using airborne laser scanning. Remote Sensing, 2012, 4 (4): 950- 974.
	Korhonen L , Korpela I , Heiskanen J , et al. Airborne discrete-return LiDAR data in the estimation of vertical canopy cover, angular canopy closure and leaf area index. Remote Sensing of Environment, 2011, 115 (4): 1065- 1080.
	Lee A C , Lucas R M . A LiDAR-derived canopy density model for tree stem and crown mapping in Australian forests. Remote Sensing of Environment, 2007, 111 (4): 493- 518.
	Li W , Guo Q , Jakubowski M K , et al. A new method for segmenting individual trees from the LiDAR point cloud. Photogrammetric Engineering & Remote Sensing, 2012, 78 (1): 75- 84.
	Liu Q W , Fu L Y , Wang G , et al. Improving estimation of forest canopy cover by introducing loss ratio of laser pulses using airborne LiDAR. IEEE Transactions on Geoscience and Remote Sensing, 2019, 99, 1- 19.
	Mei C , Durrieu S . Tree crown delineation from digital elevation models and high resolution imagery. Proc. IAPRS, 2004, 36, 218- 223.
	Parent J R , Volin J C . Assessing the potential for leaf-off LiDAR data to model canopy closure in temperate deciduous forests. ISPRS Journal of Photogrammetry and Remote Sensing, 2014, 95, 134- 145.
	Paris C , Valduga D , Bruzzone L . A hierarchical approach to three-dimensional segmentation of LiDAR data at single-tree level in a multilayered forest. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54 (7): 4190- 4203.
	Pedersen R Ø , Bollandsås O M , Gobakken T , et al. Deriving individual tree competition indices from airborne laser scanning. Forest Ecology and Management, 2012, 280, 150- 165.
	Pitkanen J . Individual tree detection in digital aerial images by combining locally adaptive binarization and local maxima methods. Canadian Journal of Forest Research, 2001, 31 (5): 832- 844.
	Reitberger J , Schnörr C , Krzystek P , et al. 3D segmentation of single trees exploiting full waveform LiDAR data. ISPRS Journal of Photogrammetry and Remote Sensing, 2009, 64 (6): 561- 574.
	Yao W , Krull J , Krzystek P , et al. Sensitivity analysis of 3D individual tree detection from LiDAR point clouds of temperate forests. Forests, 2014, 5, 1122- 1142.
	Zhang C , Zhou Y , Qiu F . Individual tree segmentation from LiDAR point clouds for urban forest inventory. Remote Sensing, 2015, 7 (6): 7892- 7913.
	Zhao K , Popescu S , Nelson R . LiDAR remote sensing of forest biomass: a scale-invariant estimation approach using airborne lasers. Remote Sensing of Environment, 2009, 113 (1): 182- 191.

样地编号 No.	树种 Tree species	类型Types	株数 Plant number	年龄 Age/a	郁闭度 Canopy density	平均树高 Mean height/m	平均胸径 Mean DBH/cm	平均冠幅 Mean canopy/m
样地1 Sample 1	油松 Pinus tabulaeformis	新造林 Newly afforestation	68	7	0.30	2.98	4.68	1.62
样地2 Sample 2	油松 Pinus tabulaeformis	新造林 Newly afforestation	85	8	0.43	3.29	4.98	1.76
样地3 Sample 3	油松 Pinus tabulaeformis	新造林 Newly afforestation	92	8	0.50	3.24	5.75	1.96
样地4 Sample 4	油松 Pinus tabulaeformis	新造林 Newly afforestation	71	7	0.37	2.39	3.68	1.18
样地5 Sample 5	油松 Pinus tabulaeformis	新造林 Newly afforestation	60	7	0.21	2.73	3.36	1.00
样地6 Sample 6	华北落叶松 Larix principis-rupprechtii	中龄林 Middle-aged forest	66	38	0.82	12.91	19.89	4.64
样地7 Sample 7	华北落叶松 Larix principis-rupprechtii	中龄林 Middle-aged forest	91	40	0.90	17.60	20.40	5.20
样地8 Sample 8	华北落叶松 Larix principis-rupprechtii	中龄林 Middle-aged forest	50	41	0.75	11.80	17.75	3.88
样地9 Sample 9	华北落叶松 Larix principis-rupprechtii	中龄林 Middle-aged forest	60	41	0.80	12.62	19.68	5.10
样地10 Sample 10	华北落叶松 Larix principis-rupprechtii	中龄林 Middle-aged forest	100	40	0.86	13.22	15.54	3.50

精度验证指标 Index for accuracy assessments	基于点云的距离判别聚类法 Point cloud-based cluster segmentation algorithm	分水岭分割算法 Watershed segmentation algorithm	双正切角树冠识别算法 Double-tangent crown recognition algorithm
r	0.95	0.87	0.89
p	0.92	0.93	0.91
F	0.93	0.90	0.90

精度验证指标 Index for accuracy assessments	基于点云的距离判别聚类法 Point cloud-based cluster segmentation algorithm	分水岭分割算法 Watershed segmentation algorithm	双正切角树冠识别算法 Double-tangent crown recognition algorithm
r	0.62	0.64	0.78
p	0.85	0.77	0.72
F	0.72	0.70	0.75

样地编号 No.	基于点云的距离判别聚类法 Point cloud-based cluster segmentation algorithm	分水岭分割算法 Watershed segmentation algorithm	双正切角树冠识别算法 Double-tangent crown recognition algorithm
样地1 Sample 1	0.815^**	0.562^**	0.798^**
样地2 Sample 2	0.828^**	0.721^**	0.825^**
样地3 Sample 3	0.866^**	0.829^**	0.819^**
样地4 Sample 4	0.911^**	0.588^**	0.793^**
样地5 Sample 5	0.936^**	0.634^**	0.684^**
样地6 Sample 6	0.701^**	0.610^**	0.782^**
样地7 Sample 7	0.736^**	0.531^**	0.726^**
样地8 Sample 8	0.680^**	0.519^**	0.765^**
样地9 Sample 9	0.687^**	0.579^**	0.710^**
样地10 Sample 10	0.661^**	0.608^**	0.752^**

[1]	朱泊东,罗洪斌,金京,岳彩荣. 高郁闭度人工林无人机激光雷达单木分割方法优化[J]. 林业科学, 2022, 58(9): 48-59.
[2]	黄洪宇,骆钰波,唐丽玉,李肖肖,彭巍,陈崇成. 基于单木位置特征的多源树木三维点云配准方法[J]. 林业科学, 2022, 58(11): 96-107.
[3]	冯林艳,谭炳香,刘清旺,周超凡,于航,张会儒,符利勇. 基于GF-2影像的崇礼冬奥核心区土地覆盖和树种分类[J]. 林业科学, 2022, 58(10): 10-23.
[4]	谢栋博,雷雅凯,张宇超,刘清旺,符利勇,陈巧. 基于机载LiDAR数据的崇礼冬奥核心区树冠覆盖率估算[J]. 林业科学, 2022, 58(10): 24-34.
[5]	张状,宗树琴,闫星蓉,张浩,黄宏超,翟跃琴,符利勇. 林分和地形因子对崇礼冬奥核心区森林地表可燃物载量的影响[J]. 林业科学, 2022, 58(10): 59-66.
[6]	张晓红,周超凡,张状,冯林艳,王利华,符利勇,谭炳香. 崇礼冬奥核心区华北落叶松人工林结构特征与优化模拟[J]. 林业科学, 2022, 58(10): 79-88.
[7]	张晓芳,郭旭展,洪亮,陈涛,符利勇,张会儒. 冬奥核心区华北落叶松和白桦单木冠幅预测模型——组级贝叶斯模型、加性模型和混合效应模型比较[J]. 林业科学, 2022, 58(10): 89-100.
[8]	张致义,辛琳,邱尔发,谢晓阳,王新宇,孙振凯,姜莎莎,赵娇. 冬奥快速通道林带四季美景度评价[J]. 林业科学, 2021, 57(12): 43-56.
[9]	霍朗宁,张晓丽. 基于机载LiDAR点云多层聚类的单木信息提取及其精度评价[J]. 林业科学, 2021, 57(1): 85-94.
[10]	李丹, 张俊杰, 赵梦溪. 基于FCM和分水岭算法的无人机影像中林分因子提取[J]. 林业科学, 2019, 55(5): 180-187.
[11]	李平昊, 申鑫, 代劲松, 曹林. 机载激光雷达人工林单木分割方法比较和精度分析[J]. 林业科学, 2018, 54(12): 127-136.