Estimation of Canopy Height is Conducted by Integrating Multi-source Remote Sensing Data from ICESat-2 and GEDI

ICESat-2光子特征参数	文件路径	描述
RH98-RH25	*/gtxx/canopy/canopy_h_metrics	每分段冠层高度百分位数包括25,50,60,70,75,80,85,90,95,98
RHmax	*/gtxx/canopy/h_max_canopy	每分段冠层高度最大值
Hmedian	*/gtxx/canopy/h_median_canopy	每分段冠层高度中位数
Coverage	*/gtxx/canopy/segment_cover	冠层覆盖百分比
Slope	*/gtxx/terrain/terrain_slope	每分段地形坡度

Table 2. Table 2Image feature parameter set

View table

View in Article

Table 2. Table 2Image feature parameter set

影像特征	参数名称	参数描述
植被指数特征	EVI	2.5(B8-B4)/(B8+6B4-7.5*B2+1)提高植被覆盖估计的稳定性
	SAVI	(B8-B4)/(B8+B4+L)*(1+L)土壤亮度影响植被指数（本文L取0.5）
	GI	(B3-B2)/(B3+B2)衡量植被绿度反映植被覆盖情况
	VSWI	(B8-B11)/(B8+B11)估计植被水分含量
	CRI	B4+(2(B5-B4)(B3-B4))估计植被叶绿素含量
	BRI	B4/B5估计植被叶绿素浓度反映植被生长状况
	CI_green	B4/B3反映植被光谱特性和生长状况
纹理特征	CON(对比度)	$\sum_{n = 0}^{k = 1} n 2 \sum_{\|i - j\| = 1} P (i, j)$ 估计植被叶绿素浓度反映植被生长状况
	ENT(熵)	$- \sum_{i = 1}^{k} \sum_{j = 1}^{k} P (i, j) l o g [(P (i, j))]$ 描述纹理的非均匀程度或复杂度的特征
	VAR(协方差)	$\sum_{i = 1}^{k} \sum_{j = 1}^{k} [P (i, j) - μ_{n + n}]^{2}$ 表示灰度的变化大小的特征
	DIS(相异性)	$\sum_{i = 1}^{k} \sum_{j = 1}^{k} \|i - j\| P (i, j)$ 描述图像灰度的相似性的特性
	CON(对比度)	$\sum_{n = 0}^{k - 1} n^{2} \sum_{\|i - j\| = 1} P (i, j)$ 描述某像素值及其邻域像素值亮度的对比情况

Table 3. Table 3The accuracy evaluation results of different models in the table

View table

View in Article

Table 3. Table 3The accuracy evaluation results of different models in the table

算法模型	训练集			测试集
算法模型	R²	RMSE/m	MAE/m	R²	RMSE/m	MAE/m
SLR	0.53	3.97	3.21	0.56	3.97	3.22
LightGBM	0.70	3.26	2.43	0.56	3.96	3.16
RF	0.73	3.08	2.45	0.64	3.49	2.85
RF-SLR	0.74	2.91	2.24	0.69	3.35	2.60
PSO-RF	0.78	2.79	2.21	0.71	3.15	2.63

Table 4. Table 4Evaluation of accuracy among different forest height dataset models
View table
View in Article
Table 4. Table 4Evaluation of accuracy among different forest height dataset models
冠层高度集训练集测试集
R² RMSE/m MAE/m R² RMSE/m MAE/m
ICESat-2 0.84 2.98 2.26 0.41 6.11 4.19
GEDI 0.62 3.77 3.02 0.38 4.76 3.75
RF-SLR 0.86 2.07 1.66 0.55 3.77 2.99
PSO-RF 0.86 1.61 1.29 0.56 3.02 2.38

Tools

Get Citation

Copy Citation Text

Huajun LIANG, Qiang BIE, Ying SHI, Xinru DENG, Xinzhang LI. Estimation of Canopy Height is Conducted by Integrating Multi-source Remote Sensing Data from ICESat-2 and GEDI[J]. Remote Sensing Technology and Application, 2025, 40(1): 202

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category:

Received: Mar. 26, 2024

Accepted: --

Published Online: May. 22, 2025

The Author Email: Qiang BIE (bieq@lzjtu.edu.cn)

DOI:10.11873/j.issn.1004-0323.2025.1.0202

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology

Table 1. Table 1ICESat-2 photon feature parameter

Table 1. Table 1ICESat-2 photon feature parameter

Table 2. Table 2Image feature parameter set

Table 2. Table 2Image feature parameter set

Table 3. Table 3The accuracy evaluation results of different models in the table

Table 3. Table 3The accuracy evaluation results of different models in the table

Table 4. Table 4Evaluation of accuracy among different forest height dataset models

Table 4. Table 4Evaluation of accuracy among different forest height dataset models