星载LiDAR与HJ-1A/HSI高光谱数据联合估测区域森林冠层高度

doi:10.11707/j.1001-7488.20160517

摘要/Abstract

摘要：

[目的] 将ICESat-GLAS波形数据与HJ-1A/HSI高光谱数据联合,借助HSI高光谱数据提供的连续高分辨率光谱信息,实现区域森林冠层高度的估测,降低由于GLAS光斑呈离散条带状分布无法覆盖整个研究区造成的估测误差。[方法] 首先,从平滑后的ICESat-GLAS波形数据中提取波形参数(波形长度W和地形坡度参数TS),基于W和TS建立GLAS森林冠层高度估测模型,并利用此模型计算研究区所有GLAS光斑内的森林冠层高度; 然后,采用最小噪声分离法(MNF)对HJ-1A/HSI高光谱数据进行降维,提取前3个MNF分量(MNF1,MNF2,MNF3); 最后,基于支持向量回归机(SVR)算法,利用GLAS估测的森林冠层高度和3个MNF分量建立区域森林冠层高度SVR估测模型,并估测研究区内无GLAS光斑覆盖区域的森林冠层高度,生成森林冠层高度分布图。[结果] 从ICESat-GLAS波形数据中提取的地形坡度参数TS与野外实测地形坡度具有显著线性关系(R²=0.78); 基于W和TS建立的GLAS森林冠层高度估测模型的R²=0.78,RMSE=2.51 m,模型验证的R²=0.85,RMSE=1.67 m; 基于支持向量回归机算法建立的SVR模型建模的R²=0.70,RMSE=3.62 m,模型验证的R²=0.67,RMSE=4.42 m。采用野外数据对最终得到的森林冠层高度分布图的估测误差进行分析,结果估测误差最大值为7.10 m,最小值为0.07 m,平均值为1.78 m,估测误差的标准差为1.49 m,Q1为0.75 m,Q3为2.31 m。[结论] 从ICESat-GLAS波形数据中提取的地形坡度参数TS能够很好地反映地形坡度的变化,本研究建立的线性关系模型可克服对数关系模型在平坦地区解释困难的问题。基于支持向量回归机算法,将ICESat-GLAS波形数据与HJ-1A/HSI高光谱数据联合,可克服ICESat-GLAS由于光斑呈离散条带状分布无法实现区域森林冠层高度估测的不足,实现对区域森林冠层高度的高精度估测。

关键词: ICESat-GLAS波形数据, HJ-1A高光谱数据, 森林冠层高度, 坡度, 支持向量回归机

Abstract:

[Objective] In this study, ICESat-GLAS waveforms was combined with HJ-1A/HSI hyperspectral imageries to realize regional estimation of forest canopy height. [Method] We extracted parameters(waveform length W and the terrain slope parameter TS)from ICESat-GLAS waveforms, and built the forest canopy height model with W and TS. The model was used to calculate forest canopy height within each GLAS footprint of the study area. For HSI imageries, the minimum noise fraction(MNF)method was applied to decrease noise and reduce the dimensionality of HSI imageries and the first three MNF components(MNF1, MNF2, MNF3)were selected for further research. Afterwards, the SVR method was applied to establish the relationship between GLAS estimated forest canopy height and the three MNF components, and accordingly the full covered regional forest canopy height map was produced. [Result] The results showed that there was a significant linear relationship between TS and terrain slope(R²=0.78). The R² and RMSE value of the forest canopy height model built by W and TS were 0.78 and 2.51 m, respectively, and the validation results were R²=0.85 and RMSE=1.67 m. The R² and RMSE of SVR model were 0.70 and 3.62 m, respectively, with the validation results of R²=0.67, RMSE=4.42 m. The estimation error of the forest canopy height map was calculated and analyzed by field sample data, and the maximum, minimum and mean value of estimation error were 7.10 m, 0.07 m and 1.78 m, respectively. The standard deviation was 1.49 m, as well as Q1 and Q3 were 0.75 m and 2.31 m, respectively. [Conclusion] TS can perfectly reflect terrain slope. In addition, the linear relationship model built in the study overcomes the difficultly explaining problem of logarithm model in flat area. The study demonstrated that it holds great potential to estimate regional forest canopy height by combining ICESat-GLAS waveforms and HJ-1A/HSI hyperspectral imageries, which overcome the disadvantage of ICESat-GLAS in the aspect of regional estimation caused by its discrete distribution and improve the estimation accuracy.

Key words: ICESat-GLAS waveforms data, HJ-1A/HSI hyperspectral imageries, forest canopy height, terrain slope, support vector regression method

中图分类号:

S771.8

邱赛, 邢艳秋, 田静, 丁建华. 星载LiDAR与HJ-1A/HSI高光谱数据联合估测区域森林冠层高度[J]. 林业科学, 2016, 52(5): 142-149.

Qiu Sai, Xing Yanqiu, Tian Jing, Ding Jianhua. Forest Canopy Height Estimation of Large Area Using Spaceborne LIDAR and HJ-1A/HSI Hyperspectral Imageries[J]. Scientia Silvae Sinicae, 2016, 52(5): 142-149.

参考文献

董立新,吴炳方,唐世浩.2011.激光雷达 GLAS 与 ETM 联合反演森林地上生物量研究.北京大学学报: 自然科学版,47(4): 703-710.
(Dong L X, Wu B F, Tang S H. 2011. Estimation of forest aboveground biomass by integrating GLAS and ETM data. ACTA Scientiarum Naturalium Universitatis Pekinensis, 47(4): 703-710.[in Chinese])
李丹,岳彩荣.2011.激光雷达在森林参数反演中的应用.测绘与空间地理信息,34(6): 54-58.
(Li D, Yue C R. 2011. The application of LiDAR in inversion of the forest parameters. Geomatics & Spatial Information Technology, 34(6): 54-58.[in Chinese])
李海涛,顾海燕,张兵,等.2007.基于MNF和SVM的高光谱遥感影像分类研究.遥感信息,(5): 12-15.
(Li H T, Gu H Y, Zhang B, et al. 2007. Research on hyperspectral remote sensing image classification based on MNF and SVM. Remote Sensing Information, (5): 12-15.[in Chinese])
马利群,李爱农.2011.激光雷达在森林垂直结构参数估算中的应用.世界林业研究,24(1): 41-45.
(Ma L Q, Li A N. 2011. Review of application of LiDAR to estimation of forest vertical structure parameters. World Forestry Research, 24(1): 41-45.[in Chinese])
庞勇,李增元,陈尔学,等.2005.激光雷达技术及其在林业上的应用.林业科学,41(3): 129-136.
(Pang Y, Li Z Y, Chen E X, et al. 2005. LiDAR remote sensing technology and its application in forestry. Scientia Silvae Sinicae, 41(3): 129-136.[in Chinese])
谭炳香.2006.高光谱遥感森林类型识别及其郁闭度定量估测研究.北京: 中国林业科学研究院博士学位论文.
(Tan B X. 2006. Extraction of forest types and estimation of forest canopy closure. Beijing: PhD thesis of Chinese Academy of Forestry.[in Chinese])
吴迪,范文义.2014.激光雷达协同多角度光学遥感数据反演树高.北京林业大学学报,36(4): 8-15.
(Wu D, Fan W Y. 2014. Forest canopy height estimation using LiDAR and optical multi-angler data. Journal of Beijing Forestry University, 36(4): 8-15.[in Chinese])
徐定.2013.基于 TM 的密云县森林郁闭度遥感估测.北京: 北京林业大学硕士学位论文.
(Xu D. 2013. Remote sensing estimation of forest canopy density in Miyun based on TM. Beijing: MS thesis of Beijing Forestry University.[in Chinese])
杨婷,王成,李贵才,等.2014.基于星载激光雷达 GLAS 和光学 MODIS 数据中国森林冠层高度制图.中国科学: 地球科学,44(11): 2487-2498.
(Yang T, Wang C, Li G C, et al. 2014. Forest canopy height mapping over China using GLAS and MODIS data. Science China: Earth Sciences, 44(11): 2487-2498.[in Chinese])
曾绍华.2006.支持向量回归机算法理论研究与应用.重庆: 重庆大学博士学位论文.
(Zeng S H. 2006. The theory research of algorithm on support vector regression and application. Chongqing: PhD thesis of Chongqing University.[in Chinese])
Anderson J E, Plourde L C, Martin M E, et al. 2008. Integrating waveform lidar with hyperspectral imagery for inventory of a northern temperate forest. Remote Sensing of Environment, 112(4): 1856-1870.
Chopping M, Nolin A, Moisen G G, et al. 2009. Forest canopy height from the multiangle imaging spectro radiometer(MISR)assessed with high resolution discrete return liDAR. Remote Sensing of Environment, 113(10): 2172-2185.
Duncanson L I, Niemann K O, Wulder M A. 2010. Estimating forest canopy height and terrain relief from GLAS waveform metrics. Remote Sensing of Environment, 114(1): 138-154.
Gwenzi D, Lefsky M A. 2014. Modeling canopy height in a savanna ecosystem using spaceborne lidar waveforms. Remote Sensing of Environment, 154: 338-344.
Hyde P, Nelson R, Kimes D, et al. 2007. Exploring LiDAR-RaDAR synergy-predicting aboveground biomass in a southwestern ponderosa pine forest using LiDAR, SAR and InSAR. Remote Sensing of Environment, 106(1): 28-38.
Iqbal I A. 2010. Evaluating the potential of ICESat/GLAS data to estimate canopy height in the New Forest National Park, UK.38^th COSPAR Scientific Assembly, 303.
Lee S, Ni-Meister W, Yang W, et al. 2011. Physically based vertical vegetation structure retrieval from ICESat data: validation using LVIS in White Mountain National Forest, New Hampshire, USA. Remote Sensing of Environment, 115(11): 2776-2785.
Lefsky M A, Harding D J, Keller M, et al. 2005. Estimates of forest canopy height and aboveground biomass using ICESat. Geophysical Research Letters, doi:10.1029/2005GL025518.
Mahoney C, Kljun N, Los S O, et al. 2014. Slope estimation from ICESat/GLAS. Remote Sensing, 6(10): 10051-11006.
Pang Y, Lefsky M, Sun G, et al. 2011. Impact of footprint diameter and off-nadir pointing on the precision of canopy height estimates from spaceborne liDAR. Remote Sensing of Environment, 115(11): 2798-2809.
Schutz B E, Zwally H J, Shuman C A, et al. 2005. Overview of the ICESat mission. Geophysical Research Letters, doi:10.1029/2005GL024009.
Simard M, Pinto N, Fisher J B, et al. 2011. Mapping forest canopy height globally with spaceborne liDAR. Journal of Geophysical Research: Biogeosciences, doi: 10.1029/2011JG001708.
Van leeuwen M, Nieuwenhuis M. 2010. Retrieval of forest structural parameters using LiDAR remote sensing. European Journal of Forest Research, 129(4): 749-770.
Xing Y, De gier A, Zhang J, et al. 2010. An improved method for estimating forest canopy height using ICESat-GLAS full waveform data over sloping terrain: a case study in Changbai mountains, China. International Journal of Applied Earth Observation, 12(5): 385-392.
Yang W, Ni-Meister W, Lee S. 2011. Assessment of the impacts of surface topography, off-nadir pointing and vegetation structure on vegetation lidar waveforms using an extended geometric optical and radiative transfer model. Remote Sensing of Environment, 115(11): 2810-2822.
Yu Y, Yang X, Fan W. 2015. Estimates of forest structure parameters from GLAS data and multi-angle imaging spectrometer data. International Journal of Applied Earth Observation, 38: 65-71.
Zhao X, Lu J, Gong A, et al. 2010. Operational spectrum reconstruction of data from the Fourier transform hyperspectral imager onboard HJ-1A satellite. Chinese Science Bulletin, 55(17): 1808-1812.

[1]	陈文思, 朱清科, 刘蕾蕾, 马欢, 赵维军, 王瑜. 陕北半干旱黄土区沙棘人工林的死亡率及适宜地形因子[J]. 林业科学, 2016, 52(5): 9-16.
[2]	范叶青, 周国模, 施拥军, 杜华强, 周宇峰, 徐小军. 地形条件对毛竹林分结构和植被碳储量的影响[J]. 林业科学, 2013, 49(11): 177-182.
[3]	王立海;潘明旭;段铁成;王秋实;胡斌. 不同坡度及作用力对伐木作业者心率及伐木锯切周期的影响[J]. 林业科学, 2012, 48(5): 173-179.