Remote Sensing Estimation of Average Diameter at Breast Height of Forest Stands Based on Airborne LiDAR and Machine Learning Algorithms

doi:10.13466/j.cnki.lczyyj.2024.01.008

Abstract

Abstract:

In order to explore the prediction accuracy of different models on the average diameter at breast height of forest stands,airborne LiDAR point cloud data and ground measured sample plot datawere obtained simultaneously by the Machang working area of Guihua State-owned Forest Farm in Guizhou Province.By extracting point cloud feature variables at the sample plot level,a machine learning model is used toestimate the average diameter at breast height of the sample plot,variance inflation factor analysis and Pearson correlation test are used to select independent variables.The results indicate that:1)Point cloud feature variables show a strong correlation with the average diameter at breast height of the forest stand,such as the average canopy height and height skewness.2)Machine learning models(random forest,support vector machine,nearest neighbor algorithm)outperform multiple linear regression models,with random forest having the best fitting performance.The determination coefficient(R²) for the random forest model is 0.71,withthe root mean square error(RMSE)of 2.50.3)The difference between the predicted and actual average diameter at breast height of four forest types:Cryptomeria forest,mixed coniferous forest,mixed coniferous and broad-leaved forest,and Pinusmassoniana forest further confirms that the random forest model has the highest accuracy and the best fitting effect.In summary,it is feasible to extract point cloud feature variables using airborne LiDAR point cloud data and construct a forest average diameter estimation model based on machine learning algorithms.The accuracy of this method meets the application requirements of forest resource investigation and can be used as a technical means to assist in forestry investigation work.

Key words: airborne LiDAR, average diameter at breast height of the forest stand, random forest, point cloud feature variables

CLC Number:

S771.8

TANG Jiajun, CHAI Zongzheng. Remote Sensing Estimation of Average Diameter at Breast Height of Forest Stands Based on Airborne LiDAR and Machine Learning Algorithms[J]. Forest and Grassland Resources Research, 2024, (1): 56-64.

Figures/Tables 9

Fig.1

Tab.1

Tab.2

Fig.2

Tab.3

Tab.4

Fig.3

Tab.5

Tab.6

References 32

[1]	李海奎, 法蕾. 基于分级的全国主要树种树高-胸径曲线模型[J]. 林业科学, 2011, 47(10):83-90.
[2]	Andersen H, McGaughey R J, Reutebuch S E. Estimating forest canopy fuel parameters using LIDAR data[J]. Remote Sensing of Environment, 2005, 94(4):441-449. doi: 10.1016/j.rse.2004.10.013
[3]	Balduzzi M A, Zande D V, Stuckens J, et al. The properties of terrestrial laser system intensity for measuring leaf geometries:A case study with conference pear trees(Pyrus Communis)[J]. Sensors, 2011, 11(2):1657-1681. doi: 10.3390/s110201657
[4]	李增元, 陈尔学. 中国林业遥感发展历程[J]. 遥感学报, 2021, 25(1):292-301.
[5]	韦雪花. 轻小型航空遥感森林几何参数提取研究[D]. 北京: 北京林业大学, 2013.
[6]	吕金城, 王振锡, 杨勇强, 等. 基于无人机影像的天山云杉林树高提取及蓄积量的反演[J]. 新疆农业科学, 2021, 58(10):1838-1845. doi: 10.6048/j.issn.1001-4330.2021.10.009
[7]	Swayze N C, Tinkham W T. Application of unmanned aerial system structure from motion point cloud detected tree heights and stem diameters to model missing stem diameters[J]. MethodsX, 2022, 9:101729. doi: 10.1016/j.mex.2022.101729
[8]	Mao Zhihui, Lu Zhuo, Wu Yanjie, et al. DBH estimation for indivi-dual tree:Two-dimensional images or three-dimensional point clouds?[J]. Remote Sensing, 2023, 15(16):4116. doi: 10.3390/rs15164116
[9]	Lefsky M A, Cohen W B, Acker S A, et al. Lidar remote sensing of the canopy structure and biophysical properties of Douglas-fir western hemlock forests[J]. Remote Sensing of Environment, 1999, 70(3):339-361. doi: 10.1016/S0034-4257(99)00052-8
[10]	Hawbaker T J, Gobakken T, Lesak A, et al. Light detection and ranging-based measures of mixed hardwood forest structure[J]. Forest Science, 2010, 56(3):313-326.
[11]	Zhang Zhengnan, Wang Tiejun, Skidmore A K, et al. An improved area-based approach for estimating plot-level tree DBH from airborne LiDAR data[J]. Forest Ecosystems, 2023, 10:100089. doi: 10.1016/j.fecs.2023.100089
[12]	Zhao Xiaoqian, Guo Qinghua, Su Yanjun, et al. Improved progressive TIN densification filtering algorithm for airborne LiDAR data in forested areas[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2016, 117:79-91. doi: 10.1016/j.isprsjprs.2016.03.016
[13]	Næsset E, Gobakken T. Estimation of above-and below-ground biomass across regions of the boreal forest zone using airborne laser[J]. Remote Sensing of Environment, 2008, 112(6):3079-3090. doi: 10.1016/j.rse.2008.03.004
[14]	Zhang Zhengnan, Cao Lin, Mulverhill C, et al. Prediction of diameter distributions with multimodal models using LiDAR data in subtropical planted forests[J]. Forests, 2019, 10(2):125. doi: 10.3390/f10020125
[15]	庞勇, 李增元. 基于机载激光雷达的小兴安岭温带森林组分生物量反演[J]. 植物生态学报, 2012, 36(10):1095-1105.
[16]	常晓敏. 华北北部防护林主要树种立地-生长-结构-功能多元耦合关系研究[D]. 北京: 北京林业大学, 2021.
[17]	Belgiu M, Drăguţ L. Random forest in remote sensing:A review of applications and future directions[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2016, 114:24-31. doi: 10.1016/j.isprsjprs.2016.01.011
[18]	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.
[19]	曹庆先. 北部湾沿海红树林生物量和碳贮量的遥感估算[D]. 北京: 中国林业科学研究院, 2010.
[20]	付逍遥. 基于机载和无人机点云数据的平原人工林结构参数动态监测[D]. 南京: 南京林业大学, 2021.
[21]	黄昕晰. 基于无人机影像与Mask R-CNN的城市单木检测与参数提取研究[D]. 杭州: 浙江农林大学, 2020.
[22]	张峥男. 杉木和桉树人工林胸径结构估测和采伐优先等级划分[D]. 南京: 南京林业大学, 2023.
[23]	Tian Dongyuan, Jiang Lichun, Shahzad M K, et al. Climate-sensitive tree height-diameter models for mixed forests in Northeastern China[J]. Agricultural and Forest Meteorology, 2022, 326:109182. doi: 10.1016/j.agrformet.2022.109182
[24]	Lines E R, Zavala M A, Purves D W, et al. Predictable changes in aboveground allometry of trees along gradients of temperature,aridity and competition[J]. Global Ecology and Biogeography, 2012, 21(10):1017-1028. doi: 10.1111/geb.2012.21.issue-10
[25]	Hulshof C M, Swenson N G, Weiser M D. Tree height-diameter allometry across the United States[J]. Ecology and Evolution, 2015, 5(6):1193-1204. doi: 10.1002/ece3.1328 pmid: 25859325
[26]	欧强新. 基于机器学习的吉林天然针阔混交林生长建模[D]. 北京: 中国林业科学研究院, 2019.
[27]	朱兆廷, 孙玉军, 梁瑞婷, 等. 基于树冠和竞争因子的杉木胸径估测[J]. 北京林业大学学报, 2023, 45(9):42-51.
[28]	陈玉玲. 人工林适地适树与生长收获效益评估研究[D]. 北京: 北京林业大学, 2020.
[29]	Mensah S, Pienaar O L, Kunneke A, et al. Height-Diameter allometry in South Africa's indigenous high forests:Assessing generic models performance and function forms[J]. Forest Ecology and Management, 2018, 410:1-11. doi: 10.1016/j.foreco.2017.12.030
[30]	Cysneiros V C, Pelissari A L, Gaui T D, et al. Modeling of tree height-diameter relationships in the Atlantic forest:Effect of forest type on tree allometry[J]. Canadian Journal of Forest Research, 2020, 50(12):1289-1298. doi: 10.1139/cjfr-2020-0060
[31]	Mokroš M, Liang Xinlian, Surovy P, et al. Evaluation of close-range photogrammetry image collection methods for estimating tree diameters[J]. ISPRS International Journal of Geo-Information, 2018, 7(3):93. doi: 10.3390/ijgi7030093
[32]	Chave J, Andalo C, Brown S, et al. Tree allometry and improved estimation of carbon stocks and balance in tropical forests[J]. Oecologia, 2005, 145:87-99. doi: 10.1007/s00442-005-0100-x pmid: 15971085

森林类型	样地数量/个	平均胸径/cm	平均树高/m	林分蓄积/m³
柳杉(Cryptomeria japonica)纯林	3	20.05	15.05	56.02
针叶混交林	7	19.84	14.35	106.03
针阔混交林	17	21.88	16.43	323.13
马尾松(Pinus massoniana)纯林	21	26.38	17.70	486.97

变量类别	变量名称	变量描述	变量
高度特征变量	1%分位数累计高度	激光返回点云1%的点所在累计高度	H_A1
	高度中位数	所有点高度值的中位数	H_m
	高度峰度	所有点高度值的平坦程度	H_k
	高度偏态	所有点高度分布的对称程度	H_S
冠层特征变量	平均冠层高度	大于高度阈值的所有点高度平均值,指树冠平均高度	I_mCH
	冠层高度标准差	林分冠层高度的标准差	I_sdCH
	郁闭度	冠层垂直投影占林地面积百分比	I_CC
	叶面积指数	叶片表面积的一半	I_LAI

变量	重要值/%	排序
I_mCH	60.94	1
I_sdCH	12.68	2
H_A1	8.90	3
H_k	7.12	4
H_m	5.48	5
H_S	4.88	6

核函数类型	决定系数(R²)	均方根误差(RMSE)/cm
线性内核	0.66	2.48
多项式内核	0.66	2.49
径向基内核	0.67	2.32

模型类型
多元线性回归(MLR)	0.56	2.98
随机森林(RF)	0.71	2.50
支持向量机(SVM)	0.67	2.32
最近邻算法(KNN)	0.65	2.31