机载激光雷达数据估测森林资源蓄积量研究

doi:10.13466/j.cnki.lyzygl.2023.01.018

摘要/Abstract

摘要：

针对传统森林资源二类调查方法周期长且费时费力,难以满足新形势下森林资源动态监测需求的问题,以南京市六合区内3个林场为研究区,利用平均点密度1点/m²的激光雷达数据提取特征变量,结合二类调查数据,使用SMLR与Boruta两种算法进行因子筛选,对比SMLR,SVM与RF这3种建模方法,估测森林蓄积量。结果表明:1)高度因子是影响森林蓄积量的主要特征参数;2)SVM和RF这算法在模型拟合与验证精度方面均表现较优,SVM算法在混交林方面表现略逊色于RF这算法,SMLR方法表现不佳。结果表明,利用激光雷达提取因子与森林蓄积量进行建模有较好的结果,稀疏性机载激光雷达对森林资源调查有较好的适用性,为今后森林资源调查提供了新的思路。

关键词: 激光雷达, 稀疏数据, 森林蓄积量估测, 森林资源调查

Abstract:

Aiming at the problem that the traditional second-class survey method of forest resources is time-consuming and laborious,it is difficult to meet the needs of forest resource dynamic monitoring under the new situation.Three forest farms in Liuhe District,Nanjing city were selected as the study area in this paper.Sparse airborne laser radar was used to extract feature parameters,combined with the data of Forest Management Inventory.Two algorithms,stepwise regression and Boruta,were used for factor screening.Three modeling methods including Stepwise Multiple Linear Regression (SMLR),Support Vector Machine (SVM) and Random Forest (RF) were compared to estimate the forest stock.The results showed that:1) Height factor was the main characteristic parameter affecting forest stock volume;2) SVM and RF algorithms performed better in model fitting and verification accuracy,SVM algorithms performed slightly inferior to RF algorithms in mixed forests,and stepwise regression methods performed poorly.In general,the modeling results of lidar extraction factor and forest stock were good,and Sparse airborne laser radar had good applicability to forest resource survey,which provided a new idea for forest resource survey in the future.

Key words: LiDAR, sparse data, forest stock estimates, forest resource survey

中图分类号:

S758

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.

QIU Jie, LI Qiannan, YU Yao. Research on Estimating Forest Stock Volume via Airborne LiDAR Data[J]. FOREST RESOURCES WANAGEMENT, 2023, 0(1): 153-160.

图/表 3

表1

模型拟合结果

树种	方法	$R C V 2$	$M A E C V$	$R M S E C V$	$r R M S E C V / %$	样本量
混交	RF	0.896	12.6	14.9	29.7	300
	SVM	0.776	10.9	20.8	41.1
	SMLR	0.400	25.0	67.6	129.7
阔叶	RF	0.911	17.0	35.9	39.3	173
	SVM	0.894	9.4	21.2	23.0
	SMLR	0.409	29.3	109.6	119.5
针叶	RF	0.900	15.0	41.5	47.1	112
	SVM	0.884	7.8	22.7	25.9
	SMLR	0.617	21.9	75.2	85.1

表1

表2

模型整体验证精度

树种	方法	$R C V 2$	$M A E C V$	$R M S E C V$	$r R M S E C V / %$	样本量
混交	RF	0.575	23.2	84.1	169.8	75
	SVM	0.506	25.7	107.7	207.7
	SMLR	0.439	25.9	115.0	221.0
阔叶	RF	0.406	31.7	217.3	237.2	43
	SVM	0.425	33.1	257.3	281.0
	SMLR	0.334	34.5	293.5	318.9
针叶	RF	0.541	23.8	188.1	211.9	27
	SVM	0.491	28.1	262.5	299.0
	SMLR	0.402	31.0	300.2	343.7

表2

图1

参考文献 20

[1]	刘恒. 基于多源数据的森林资源二类调查蓄积量估测研究[D]. 西安: 西安科技大学, 2019.
[2]	戴永江. 激光雷达原理[M]. 北京: 国防工业出版社, 2002.
[3]	庞勇, 李增元, 陈尔学, 等. 激光雷达技术及其在林业上的应用[J]. 林业科学, 2005(3):129-136.
[4]	袁钰娜, 彭道黎, 王威, 等. 利用机载激光雷达技术估测东北林区典型针叶林的蓄积量[J]. 应用生态学报, 2021, 32(3):836-844. doi: 10.13287/j.1001-9332.202103.001
[5]	洪奕丰, 林辉, 严恩萍, 等. 基于偏最小二乘法的平南县森林蓄积量估测模型研究[J]. 中南林业科技大学学报, 2011, 31(7):80-85.
[6]	孙忠秋, 高金萍, 吴发云, 等. 基于机载激光雷达点云和随机森林算法的森林蓄积量估测[J]. 林业科学, 2021, 57(8):68-81.
[7]	杜志, 陈振雄, 马开森, 等. 基于无人机LiDAR特征变量的南方集体林区蓄积量估测[J]. 林业资源管理, 2021(4):157-165.
[8]	曾伟生, 孙乡楠, 王六如, 等. 基于机载激光雷达数据的森林蓄积量模型研建[J]. 林业科学, 2021, 57(2):31-38.
[9]	Kaartinen H, Hyyppa J, Yu X, et al. An international comparison of individual tree detection and extraction using airborne laser scanning[J]. Remote Sensing, 2012, 4:950-974. doi: 10.3390/rs4040950
[10]	Yu X, Hyypp J, Kukko A, et al. Change detection techniques for canopy height growth measurements using airborne laser scanner data[J]. Photogrammetric Engineering & Remote Sensing, 2015, 72(72):1339.
[11]	Ferreiro G, Aranda D, Miranda A. Estimation of stand variables in Pinus radiata D.Don plantations using different LiDAR pulse densities[J]. Forestry, 2012, 85(2):281-292. doi: 10.1093/forestry/cps002
[12]	Tesfamichael S G, Beech C. Combining Akaike's information criterion and discrete return LiDAR data to estimate structural attributes of savanna woody vegetation[J]. Journal of Arid Environments, 2016, 129:25-34. doi: 10.1016/j.jaridenv.2016.02.006
[13]	魏金龙, 李明阳, 赵邑晨, 等. 基于稀疏型机载激光雷达数据的风景林参数估测[J]. 西北林学院学报, 2021, 36(2):164-171.
[14]	黄宇玲, 吴达胜, 方陆明. 基于逐步回归的XGboost方法的森林蓄积量估测[J]. 中南林业科技大学学报, 2020, 40(6):72-80.
[15]	王平. 基于改进Boruta算法的森林地上生物量遥感估测研究[D]. 长沙: 中南林业科技大学, 2022.
[16]	林卓, 吴承祯, 洪伟, 等. 基于BP神经网络和支持向量机的杉木人工林收获模型研究[J]. 北京林业大学学报, 2015, 37(1):42-47.
[17]	Burges C J C. A tutorial on support vector machines for pattern recognition[J]. Data Mining and Knowledge Discovery, 1998, 2(2):121-167. doi: 10.1023/A:1009715923555
[18]	李紫荆, 刘彦枫, 冉娇. 基于Landsat 8 OLI的云南元江流域森林生物量光学遥感估测及其饱和点分析[J]. 西南林业大学学报:自然科学, 2023, 43(1):126-136.
[19]	刘海, 苏本跃. 基于激光雷达数据的森林蓄积量模型反演及精度估算[J]. 安徽工程大学学报, 2021, 36(4):34-40.
[20]	Nsset E. Practical large-scale forest stand inventory using a small footprint airborne scanning laser[J]. Scandinavian Journal of Forest Research, 2004, 19(2):164-179. doi: 10.1080/02827580310019257