文献类型：期刊文献

中文题名：基于一种简单物理模型的叶面积指数反演

英文题名：Inversion of leaf area index based on a simple physical model

作者：王强[1,2] 庞勇[3] 李增元[3] 陈尔学[3] 孙国清[4] 谭炳香[3] 刘丹丹[2]

第一作者：王强

通信作者：Wang, Qiang

机构：[1]哈尔滨工业大学电子与信息工程学院,黑龙江哈尔滨150001;[2]黑龙江工程学院测绘工程学院,黑龙江哈尔滨150050;[3]中国林业科学研究院资源信息研究所,北京100091;[4]马里兰大学地理系,马里兰20742

年份：2016

卷号：45

期号：3

起止页码：623-629

中文期刊名：中国矿业大学学报

外文期刊名：Journal of China University of Mining & Technology

收录：CSTPCD;;EI(收录号：20162202445250);Scopus(收录号：2-s2.0-84969972405);北大核心:【北大核心2014】；CSCD:【CSCD2015_2016】；

基金：国家自然科学基金项目(41201435);国家重点基础研究发展计划(973)项目(2013CB733404);空间地理信息综合实验室开放基金项目(KJKF-12-02,KJKF-14-02)

语种：中文

中文关键词：叶面积指数;混合象元;经验模型;物理模型;定量遥感

外文关键词：leaf area index; mixed pixel; empirical model; physical model; quantitative remotesensing

分类号：P237

摘要：叶面积指数(LAI)是植被冠层结构最基本的参量之一,也是植被定量遥感的重要参数,反演方法主要有经验模型、半经验模型与物理模型.提出了一种基于四分量的简单物理模型,能快速准确地模拟森林冠层反射率并建立查找表,以Landsat-7ETM+遥感图像中单个像元作为测算单位,对大野口典型森林区叶面积指数进行反演,利用研究区地面实测数据对模型反演结果进行验证和精度分析.结果表明:相比二分量物理模型,改进的四分量物理模型能较好地描述遥感图像植被覆盖混合像元真实的物理组成,提高冠层光谱曲线模拟精度,使LAI反演准确度有所提高,R2(R为相关系数)由0.240 3提高到0.772 7,均方根误差由0.983 0降低到0.114 8,说明四分量模型能真实地反映研究区叶面积指数分布状况.
The leaf area index （LAD is one of the basic parameters of canopy structure. It is es- sential to the quantitative remote sensing research on vegetation, and can be deduced by empiri- cal model, semi-empirical model and physical models. In this paper, a simple four-component physical model was proposed to simulate the forest canopy reflectance. A look up table （LUT） was built and combined with Landsat-7 ETM＋ image to inverse the LAI in the study area; ground measured LAI was used to compare with the LAI inverse＆ Results show that the modi- fied physical model can more accurately describe the components in mixed pixels of remote sensing image than previous two-component model. The canopy spectral simulation accuracy was improved to obtain higher retrieval LAI preciseness. The correlation coefficient R2 was im-proved from 0. 240 3 to 0. 772 7, and the root-mean-square error was reduced from 0. 983 0 to 0. 114 8, suggesting that the new model can authentically describe LAI distribution in the study area.