无人机密集匹配点云与机载激光雷达点云的差异分析

doi:10.13466/j.cnki.lyzygl.2020.03.011

林业资源管理 ›› 2020›› Issue (3): 58-62.doi: 10.13466/j.cnki.lyzygl.2020.03.011

无人机密集匹配点云与机载激光雷达点云的差异分析

林鑫¹^,²(), 庞勇², 李春干¹()

1.广西大学林学院,南宁 530004
2.中国林业科学研究院资源信息研究所,北京 100091

收稿日期:2020-04-27 修回日期:2020-06-02 出版日期:2020-06-28 发布日期:2020-07-30
通讯作者: 李春干
作者简介:林鑫(1995-),男,广西南宁人,在读硕士,主要研究方向:生态遥感。Email: xxfxhp@126.com
基金资助:
国家重点研发计划“多尺度落叶松人工林生长预测”(2017YFD0600404);科技部中国-捷克交流项目“基于无人机遥感的中幼林参数估测”(2019-43-16)

Analysis of the Deference Between the UAV Dense Matching Point Cloud and Airborne LiDAR Point Cloud

LIN Xin¹^,²(), PANG Yong², LI Chungan¹()

1. Collegeof Forestry,Guangxi University,Nanning,530005,China
2. Institute of Forest Resource Information Techniques,Chinese Academy of Forestry,Beijing,100091,China

Received:2020-04-27 Revised:2020-06-02 Online:2020-06-28 Published:2020-07-30
Contact: Chungan LI

摘要/Abstract

摘要：

为分析无人机密集匹配点云和机载激光雷达点云的异同性,对密集林分(郁闭度0.85)、稀疏林分(郁闭度0.55)和未成林地的2种点云的空间分布进行目视对比分析,并通过2种点云生产的DEM(UAV_DEM和LiDAR_DEM)分别对密集匹配点云进行归一化处理,得到2套归一化密集匹配点云数据,将其与激光雷达点云进行统计特征参数配对样本t检验分析。结果表明:1)在密集林分中,密集匹配点云无法获取冠层内部和地面信息,采用LiDAR_DEM进行归一化后,密集匹配点云的中下部分位数高度及全部分位数密度与激光雷达点云相应统计特征参数均存在显著性差异(α=0.05),但中上层分位数高度的差异不显著;2)在稀疏林分和未成林地中,除下部分位数高度外,其余高度、密度统计特征参数均与激光雷达点云相应参数无显著性差异,但密集匹配点云对幼树三维结构的刻画能力优于机载激光雷达点云。在森林调查监测中,无人机密集匹配点云可直接用于稀疏林分和未成林地的森林参数估测,在既有高精度DEM支持下可对密集林分的一些林分参数(如冠层表面高度等)进行估测。

关键词: 点云, 密集匹配, 无人机, 激光雷达, 林分密度, 森林资源调查

Abstract:

In order to analyze the similarities and differences between the UAV dense matching point cloud and the airborne LiDAR point cloud,the spatial distribution of two kinds of point clouds for dense forest(canopy density of 0.85),sparse forest(canopy density of 0.55) and undeveloped forest was visually analyzed.The densely matched point cloud was normalize through two types of DEM(UAV_DEM and LiDAR_DEM) produced by UAV and LiDAR point clouds respectively.Then,the statistical characteristics of the obtained two sets of normalized densely matched point cloud data and LiDAR point cloud data were compared using the paired sample t-test analysis.The results show that:1) In dense forests,densely matched point cloud has great limitations to obtain canopy internal and ground information.After normalization using LiDAR_DEM,the densely matched point cloud and the laser point cloud were examined with significant differences in the middle and lower quantile heights and all density characteristics(α=0.05),while no significant differences found in the middle and upper quantile heights;2) In sparse forest and undeveloped forest,except for the lower quantile height,the remaining statistical parameters of height and density are not significantly different between the densely matched point cloud and the laser point cloud.However,the densely matched point cloud is superior to the airborne LiDAR point cloud regarding the ability of describing the three-dimensional structure of young trees.In conclusion,the UVA dense matching point cloud can be directly used to estimate the forest parameters of sparse forest and undeveloped forest in the forest survey and monitoring.With the assistance of high-precision DEM,some parameters of dense forest(e.g.,crown layer height) can be further estimated through the UVA dense matching point cloud.

Key words: point cloud, dense matching, Unmanned Aerial Vehicle, LiDAR, stand density, forest inventory

中图分类号:

S758:S771.5

林鑫, 庞勇, 李春干. 无人机密集匹配点云与机载激光雷达点云的差异分析[J]. 林业资源管理, 2020,(3): 58-62.

LIN Xin, PANG Yong, LI Chungan. Analysis of the Deference Between the UAV Dense Matching Point Cloud and Airborne LiDAR Point Cloud[J]. FOREST RESOURCES WANAGEMENT, 2020,(3): 58-62.

图/表 2

参考文献 13

[1]	Wulder M A, White J C, Nelson R F, et al. Lidar sampling for large-area forest characterization:a review[J]. Remote Sensing of Environment, 2012,121:196-209. doi: 10.1016/j.rse.2012.02.001
[2]	Li Dan, Gu Xingfa, Pang Yong, et al. Estimation of forest aboveground biomass and leaf area index based on digital aerial photograph data in Northeast China[J]. Forests, 2018,9(5):275.doi: 10.3390/f9050275 doi: 10.3390/f9050275
[3]	White J C, Wulder M A, Vastaranta M, et al. The utility of Image-based point clouds for forest inventory:A comparison with airborne laser scanning[J]. Forests, 2013,4(3):518-536. doi: 10.3390/f4030518
[4]	Bohlin J, Wallerman J, Fransson J E S, Forest variable estimation using photogrammetric matching of digital aerial images in combination with a high-resolution DEM[J]. Scand J For Res, 2012,27:692-699. doi: 10.1080/02827581.2012.686625
[5]	Zarco-Tejada P J, Diaz-Varela R, Angileri V, et al. Tree height quantification using very high resolution imagery acquired from an unmanned aerial vehicle(UAV) and automatic 3D photo-reconstruction methods[J]. European Journal of Agronomy, 2014,55(2):89-99. doi: 10.1016/j.eja.2014.01.004
[6]	李亚东, 冯仲科, 明海军, 等. 无人机航测技术在森林蓄积量估测中的应用[J]. 测绘通报, 2017(4):66-69.
[7]	Fritz A, Kattenborn T, Koch B. UAV-based photogrammetric point clouds—Tree stem mapping in open stands in comparison to terrestrial laser scanner point clouds[J]. Int.Arch.Photogramm.Remote Sens.Spat.Inf.Sci., 2013,40:141-146.
[8]	曹忠. 低空无人机森林资源调查与监测应用现状与展望[J]. 林业建设, 2016(6):1-5.
[9]	Pang Yong, Li Zengyuan, Ju Hongbo, et al. LiCHy:the CAF's Li DAR,CCD and hyperspectral integrated airborne observation system[J]. Remote sensing, 2016,8(5):398. doi: 10.3390/rs8050398
[10]	Lucieer A, Darren T, King D H, et al. Using an Unmanned Aerial Vehicle(UAV) to capture micro-topography of Antarctic moss beds[J]. International Journal of Applied Earth Observation and Geoinformation, 2014,27:53-62. doi: 10.1016/j.jag.2013.05.011
[11]	Lisein J, Pierrot-Deseilligny M, Bonnet S, et al. A Photogrammetric Workflow for the Creation of a Forest Canopy Height Model from Small Unmanned Aerial System Imagery[J]. Forests, 2013,4:922-944. doi: 10.3390/f4040922
[12]	Hemandez J G, Ferreiro E G, Samento A, et al. Using high resolution UAV imagery to estimate tree variables in Pinus pinea plantation in Portugal[J]. Forest Systems, 2016,25(2):9.
[13]	Ullah S, Dees M, Datta P, et al. Comparing Airborne Laser Scanning,and Image-Based Point Clouds by Semi-Global Matching and Enhanced Automatic Terrain Extraction to Estimate Forest Timber Volume[J]. Forests, 2017,8(6):215.DOI: 10.3390/f8060215. doi: 10.3390/f8060215

归一化方式	林分	统计指标	高度指标/m						密度指标/%
归一化方式	林分	统计指标	ph25	ph50	ph75	Hmean	Hmax		dh25	dh50	dh75	CC
UAV_DEM	密集林分	均值	—	—	—	—	—	—		—	—	—
	密集林分	标准差	—	—	—	—	—	—		—	—	—
	稀疏林分	均值	3.81^**	1.62 NS	1.32 NS	1.52 NS	0.85 NS	-11.29 NS		-19.97 NS	-10.91 NS	-7.78 NS
	稀疏林分	标准差	3.91	2.44	2.27	2.15	2.31	17.68		14.67	17.73	4.88
	未成林地	均值	-0.59^*	-0.31 NS	-0.29 NS	-0.21 NS	-0.18 NS	-22.13 NS		-56.63^**	-63.81^***	—
	未成林地	标准差	0.33	0.19	0.17	0.16	0.14	21.46		38.86	45.23	—
	密集林分
	密集林分	LiDAR_DEM	均值	6.57^***	6.83^***	1.27 NS	1.72 NS	1.53 NS	65.81^*		61.88^*	62.93^**	13.56 NS
标准差			2.59	3.17	1.66	1.82	1.5	35.6		35.57	35.54	5.41
标准差	稀疏林分		2.59	3.17	1.66	1.82	1.5	35.6		35.57	35.54	5.41
均值			4.54^*	2.40 NS	1.33 NS	1.48 NS	1.39 NS	-15.56 NS		-12.01 NS	-11.73 NS	-7.32 NS
标准差			3.63	2.71	2.48	2.41	2.46	15.13		14.8	14.83	4.56
标准差	未成林地		3.63	2.71	2.48	2.41	2.46	15.13		14.8	14.83	4.56
均值			-0.47^*	-0.29 NS	-0.31 NS	-0.26 NS	-0.13 NS	-60.12^**		-65.33^***	-72.63^***	—
标准差			0.28	0.13	0.18	0.14	0.12	22.37		26.57	39.3	—

[1]	唐佳俊, 柴宗政. 基于机载激光雷达和机器学习的林分平均胸径遥感估测[J]. 林草资源研究, 2024, 0(1): 56-64.
[2]	朱赞, 王勇军, 王建琦, 许玉兰, 丘鑫琪, 万喜. 基于无人机倾斜摄影的广西桉树冠层碳储量计量模型构建[J]. 林草资源研究, 2024, 0(1): 88-94.
[3]	唐佳俊, 汪刚, 柴宗政. 基于机载激光雷达点云数据的马尾松单木材积估测[J]. 林草资源研究, 2023, 0(6): 105-112.
[4]	孟先进, 林寿明, 秦琳, 黄宁辉, 丁胜, 薛亚东, 罗勇, 杨廷栋. 绿美广东生态建设示范区数字孪生应用研究[J]. 林草资源研究, 2023, 0(5): 113-121.
[5]	邢元军, 温坤剑, 郭晓妮, 宋亚斌, 胡中岳, 江腾宇, 贺紫荆. 基于Mask R-CNN的平原造林地区单木树冠分割[J]. 林业资源管理, 2023, 0(3): 115-120.
[6]	罗丹, 王庆飞, 晁碧霄, 李乐, 郝泽周, 卢渊, 王成, 武瑞琛, 刘飞鹏, 裴男才. 广州市森林和城镇交界域林分尺度上的火险等级评价[J]. 林业资源管理, 2023, 0(3): 56-64.
[7]	曹庆安, 左勇, 何原荣, 冷鹏. 基于激光点云数据的森林单木位置提取研究[J]. 林业资源管理, 2023, 0(1): 133-140.
[8]	邱洁, 李倩楠, 虞瑶. 机载激光雷达数据估测森林资源蓄积量研究[J]. 林业资源管理, 2023, 0(1): 153-160.
[9]	徐干君, 王亚萍, 李鑫洋, 郭楠, 杨玫, 贺丹. 基于UAV航测的等级道路绿地景观多样性探讨[J]. 林业资源管理, 2023, 0(1): 161-172.
[10]	张信杰, 郑焰锋, 温坤剑, 王鹏杰, 吴发云. 融合机载和背包激光雷达的桉树单木因子估测[J]. 林业资源管理, 2022, 0(6): 131-137.
[11]	王补, 谭伟, 王贵林, 蒲秀青. 基于无人机多光谱影像的松材线虫病单木尺度监测[J]. 林业资源管理, 2022, 0(5): 107-117.
[12]	李苏春, 林露花, 夏磊, 胡璐璐, 徐小军. 基于改进的局部最大值法提取杉木单木位置[J]. 林业资源管理, 2022, 0(5): 60-68.
[13]	谢贤胜, 苏宏新, 杨元征, 李春海, 卢峰, 罗蔚生, 徐占勇. 基于地基激光雷达估测广西桉树人工林的林分参数[J]. 林业资源管理, 2022, 0(2): 100-108.
[14]	李建德, 解雅麟, 冯亮, 张致铭, 雷相东. 星站差分定位在全国森林资源调查中的应用[J]. 林业资源管理, 2022, 0(2): 39-44.
[15]	付安民, 高显连, 吴发云, 高金萍. 机载激光雷达树高产品的释义与精度水平评估[J]. 林业资源管理, 2022, 0(1): 114-123.

无人机密集匹配点云与机载激光雷达点云的差异分析

Analysis of the Deference Between the UAV Dense Matching Point Cloud and Airborne LiDAR Point Cloud

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 2

参考文献 13

相关文章 15

编辑推荐

Metrics

本文评价