基于多角度遥感物理模型的叶面积指数反演

Retrieval of conifer LAI based on the multiple-angle model

与传统遥感观测相比,多角度对地观测通过对林木多个方向的观察,可得到丰富的森林三维空结构信息,为定量遥感提供新的途径。本文根据不同遥感数据选择合适的模型并建立相应的查找表。TM/ETM+数据采用混合像元分解模型,CHRIS数据考虑不同森林场景选择不同的遥感物理模型,在选择合适的模型基础上,根据模型的不同敏感参数和试验区选择的特点设计查找表参数,并由模型正演建立查找表,再根据插值的方法由遥感图像的反射率值反演LAI。两种数据反演得到的LAI与MODIS 15A2 LAI产品比较分析。结果表明多角度遥感反演准确率有一定的提高,具有一定的理论研究和实际应用价值。

Compared with the traditional remote sensing, multi-angle obersvation could get the detailed three-dimensional structural information of forest through different directional observations, which provides a new method for quantitative remote sensing. In the paper look-up table is constructed according with the different kinds of remote sensing model. The mixed pixel model is used as TM/ETM ＋ data is used to retrieval, and the appropriate physical model is used according to the different forest sense as CHRIS data is used. Base on the model the look-up tables were constructed, and the leaf area index is received from the different kinds of data making use of interpolation. Compare the LAI with MODIS 15A2 LAI, the result showes that the multi-angle remote sensing can improve the precision of retrieval of conifer LAI, and there is a certain theoretical research and practical application value.