基于散射机制分解的ESPRIT植被高度反演新方法

ERPRIT算法估计的地表散射相位由于易受到森林去极化成分的影响而出现较大的偏差,从而导致植被高度反演结果严重偏低.针对这一问题,本文提出一种基于散射机制分解的ESPRIT植被高度反演算法.该算法关键点是利用Freeman分解理论和极化干涉互协方差矩阵估计得到更加准确的地表散射相位,以替代ESPRIT算法估计的地表散射相位,进而改善植被高度反演的精度.最后,分别利用欧空局（ESA）免费发布的PolSARpro软件模拟L波段极化干涉合成孔径雷达数据和PALSAR真实星载数据验证改进算法的有效性.

AnNew Vegetation Height Estimation Method Based on Scattering Mechanism Decomposition and ESPRIT Algorithm

Due to the ground interferometric phase is affected by forest depolarized components, vegetation height inversion results based on ESPRIT algorithm are lower than real height. In order to solve this problem, a modified ESPRIT algorithm based on scattering mechanism decomposition is proposed in this paper. The key point of this method is to use the more accurate ground interferometric phase based on Freeman decomposition concept and polarimetric interferometric crosscovariance matrices rather than that of traditional ESPRIT algorithm. The proposed algorithm can get more accurate inversion results. Finally, to validate the modified algorithm, simulated L-band PolInSAR data obtained by the PolSARProSim software of ESA and the real PALSAR data are tested.