融合遥感先验信息的叶面积指数反演  

Retrieval of Leaf Area Index by Fusing Prior Information from Remote Sensing Data

作　　者：徐卫星 薛华柱[1] 靳华安[2] 李爱农[2] Xu Weixing;Xue Huazhu;Jin Huaan;Li Ainong(Henan Polytechnic University,School of Surveying and Land Information,Engineering,Jiaozuo 454150,China;Institute of Mountain Hazards and Environment,Chinese Academy of Science,Chengdu 610041,China)

机构地区：[1]河南理工大学测绘与国土信息工程学院,河南焦作454150 [2]中国科学院水利部成都山地灾害与环境研究所,四川成都610041

出　　处：《遥感技术与应用》2019年第6期1235-1244,共10页Remote Sensing Technology and Application

基　　金：国家自然科学基金面上项目(41671376);国家自然科学基金重点项目(41631180);国家自然科学基金青年项目(41301385)

摘　　要：借助植被辐射传输模型,利用遥感观测数据估算LAI是一种较为可靠和稳健的反演方法。然而,地表的复杂性、遥感观测的有限性以及自相关性导致遥感数据包含的信息量不足,不能完全支持LAI等地表参数的估算,易造成“病态”反演。在遥感反演过程中引入先验知识能够有效地解决该问题。研究基于遥感数据提取LAI先验信息,并将其用于代价函数的构建,利用PROSAIL辐射传输模型和遗传算法,分别在500 m和250 m尺度反演LAI。将高空间分辨率LAI分别升尺度到500 m和250 m,验证对应尺度LAI结果,评价引入先验信息对于提高LAI反演精度的作用。研究表明,引入先验信息有助于提高不同分辨率下LAI反演精度,且先验信息的质量一定程度上也影响着LAI反演结果。与未加入先验信息的LAI反演结果相比,以MODIS LAI产品作为先验信息反演的500 m尺度LAI结果精度R2由0.55提高至0.65,RMSE由1.29下降至0.38。在250 m尺度,以500 m LAI反演结果作为先验信息反演的叶面积指数,其精度优于以MODIS LAI产品为先验知识的估算结果,验证精度R2增加了0.08,RMSE减少了0.18。研究使用的先验信息主要来自遥感数据本身,没有地面实测数据的参与,在此基础上发展的多分辨率LAI反演方法具有估算大区域尺度LAI的应用潜力。The estimation of leaf area index using remote sensing observation data depend on canopy radiative transfer models is a reliable and robust method.However,the information deficiency contained in the remote sensing data derived from the limitations of the surface heterogeneity,remote sensing observation and self-cor⁃relation,which cannot fully support the retrieval of surface parameters(e.g.LAI)and easily bring about the re⁃trieval become ill-posed.The problem can be solved or alleviated effectively by introducing prior knowledge.This paper come up with an approach to extract priori information of Leaf Area Index(LAI)from the remote sensing data,and the information is utilized to construct cost function,PROSAIL radiative transfer model and genetic algorithm are coupled to retrieve LAI at 500 m and 250 m scales.Then the 10 m spatial resolution LAI is upscaled to 500 m and 250 m respectively to verify the corresponding LAI result,and evaluate effects of intro⁃duction of prior information on improving LAI accuracy.The comparison of performance between LAI result us⁃ing MODIS LAI as prior information at 500 m scale with one without prior information indicates that R2 in⁃creased from 0.55 to 0.65 and RMSE decreased from 1.29 to 0.38.The LAI result using 500 m optimal LAI re⁃sult as prior information at 250 m scale is better than the estimation result with MODIS LAI priori knowledge,verification result shows that R2 increased by 0.08,RMSE decreased by 0.18.It is shown that LAI retrieval ac⁃curacy can be enhanced by auxiliary of LAI prior information,besides prior information quality also affects the LAI result to some extent.Multi-resolution LAI retrieval method developed in this paper has potential to esti⁃mate spatial and temporal LAI on large scale.

关 键 词：叶面积指数 遥感反演 先验信息 MODIS 

分 类 号：P237[天文地球—摄影测量与遥感] Q948.1[天文地球—测绘科学与技术]