联合TanDEM-X DEM与Sentinel-2多光谱数据的林下地形提取  

Sub-canopy topography extraction via TanDEM-X DEM combined with Sentinel-2 multispectral data

作　　者：刘志卫 赵蓉 朱建军[1] 付海强[1] 周璀[3] 周亦 LIU Zhiwei;ZHAO Rong;ZHU Jianjun;FU Haiqiang;ZHOU Cui;ZHOU Yi(School of Geosciences and Info-Physics,Central South University,Changsha 410083,China;School of Civil Engineering,Central South University of Forestry and Technology,Changsha 410004,China;College of Science,Central South University of Forestry and Technology,Changsha 410004,China)

机构地区：[1]中南大学地球科学与信息物理学院,长沙410083 [2]中南林业科技大学土木工程学院,长沙410004 [3]中南林业科技大学理学院,长沙410004

出　　处：《遥感学报》2024年第10期2540-2552,共13页NATIONAL REMOTE SENSING BULLETIN

基　　金：国家自然科学基金(编号:42104016,41820104005,42074016,42030112);湖南省研究生创新项目(编号:CX20210278);中南大学研究生创新项目(编号:2021zzts0260);国家留学基金委博士联合培养项目(编号:202106370125)。

摘　　要：针对植被覆盖区TanDEM-X DEM无法描述精细化林下地形的问题,本文提出了一种联合TanDEM-X DEM和Sentinel-2多光谱数据的林下地形提取方法。首先,将TanDEM-X DEM和Sentinel-2的多波段信息作为输入变量、高精度林下地形数据(LVIS测高数据)作为输出变量,通过随机森林拟合方法构建林下地形预测模型;之后,利用得到的训练模型实现无参考数据区域的林下地形提取。为了验证本文提出的方法,选择位于非洲加蓬的两个典型试验区进行验证。结果表明:提出的方法能够有效地较正TanDEM-X DEM中包含的森林高度偏差,同时提取更为精细的林下地形信息;相较于原始TanDEM-X DEM,本文方法所提取的地形精度在两个试验区分别提升了76%和63%;此外,本文方法生成的林下地形结果保持了较为完整的地形纹理,可以较好的描述林下地形细节。因此,本研究为采用TanDEM-X DEM获取大范围林下地形,提供了一种可行的方案。The Digital Elevation Model(DEM)is one of the most important data sources for various scientific studies and applications.Currently,one important data source for large-scale DEM generation originates from the TerraSAR-X add-on for digital elevation measurement(TanDEM-X)mission,which provides bistatic interferometric Synthetic Aperture Radar(InSAR)data with high spatial resolution(12 m)at the global scale.However,in forest areas,the retrieval of the subcanopy topography using TanDEM-X InSAR data still faces notable challenges because of the effects of the forest scattering process on InSAR height measurements and the limited penetration capability of X-band’s signals,causing the measured elevation to be between the ground surface and the top of the tree canopy.Although SAR signals with long wavelength has strong penetrability in the forest layer,subcanopy topography still cannot be measured due to the volume scattering effect from tree canopies or trunks.In addition,the missing space-borne PolInSAR or TomoSAR data pose another limitation for subcanopy topography estimation.In this study,a new method to extract subcanopy topography over forested areas is proposed.The method uses a combination of TanDEM-X DEM and Sentinel-2 multispectral data.TanDEM-X DEM and the multiband data of Sentinel-2 are regarded as the input variables,while the high-precision ground elevation data was considered as the target variable.Subsequently,the random forest fitting method is used to construct the subcanopy topography estimation predictive model.According to the obtained model,we can extract a large-scale subcanopy topography over the areas without reference data.Results show that the subcanopy topography derived via the proposed method has an RMSE of 3.7 and 7.78 m for the two forest sites,representing an improvement of approximately 76%and 63%,respectively,in comparison with the original TanDEM-X DEM.The experimental results also show that the resultant subcanopy topography can maintain more detailed topographic information.All t

关 键 词：遥感 TanDEM-X Sentinel-2 机器学习 数字高程模型 林下地形 

分 类 号：P237[天文地球—摄影测量与遥感] P2[天文地球—测绘科学与技术]