基于极化分解的极化特征参数提取与应用  

Extraction and Application of Polarimetric Characteristic Parameters Based on Polarimetric Decomposition

作　　者：王庆[1] 曾琪明[1] 廖静娟[2] 

机构地区：[1]北京大学遥感与GIS研究所,北京100871 [2]中国科学院对地观测与数字地球科学中心,北京100094

出　　处：《国土资源遥感》2012年第3期103-110,共8页Remote Sensing for Land & Resources

基　　金：国家863课题(编号:2012AA121304;2008AA121806);教育部博士点基金(编号:20090001110039)共同资助

摘　　要：基于极化分解原理,获取了描述地物散射机制的特征参数,并组合成一些特征指数,如雷达植被指数等。这些特征指数具有反映体散射信息的能力,从而可间接获取植被长势、疏密程度及分布区域等信息。实验选择了鄱阳湖区Radarsat-2全极化数据,结合野外采集的样本数据,在分析该区植被特征的基础上,对不同特征参数进行了对比分析,对雷达植被指数与实地测量样本的生物量参数进行了相关分析。实验结果表明:文中给出的4种特征参数对植被引起的随机散射的描述总体趋势是一致的,但随着植被覆盖密度的增大,不同特征指数具有一定的差异,其中雷达植被指数最为准确,适用动态范围最大,并且与湿地植被生物量具有较高的线性相关性,可以定量地反映研究区的植被疏密及生物量差异信息。Based on the polarization decomposition, the authors obtained a set of characteristic parameters to characterize the surface features, and formed some indices by combing these characteristic parameters to extract vegetation growing information such as radar vegetation indexes. These characteristic parameters have the function of reflecting the backscattering information and indirectly obtaining such information as vegetation growing, density and distribution areas. A scene of Radarsat -2 full polarimetric SAR data covering Poyang Lake region was chosen for this experiment and the sample data which included the biomass and vegetation density information were collected in the field at the same time. On the basis of an analysis of vegetation characteristics in this test region, the characteristic parameters were compared with each other and analyzed for their physical meaning. The radar vegetation index and field measurements of biomass sample parameters were statistically correlated. The experimental results show that the four characteristic parameters described in this paper give the same overall trend on the random scattering of vegetation, but different indexes have different indications with the increasing vegetation density, of which the most accurate index is the radar vegetation index, which has the maximum dynamic range. The radar vegetation index has a high linear correlation with the biomass of wetland vegetation, so that it can be used to quantitatively infer the vegetation covering density and the biomass information.

关 键 词：非相干分解 极化分解 散射机制 雷达植被指数 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]