树高是森林资源的重要参数,因此对于树高反演的研究具有重要的意义。以欧空局提供的PolSARPro软件生成L波段和X波段的全极化雷达数据,对2个波段的数据进行预处理,选择DEM差值法、RVOG地相位法和复相干幅度法对树高进行反演,并对L波段和X波段反演结果进行分析。结果表明:对于L波段,DEM差值法低估了树高,RVOG地相位法高估了树高,复相干幅度法反演的树高精度比DEM差值法和RVOG地相位法要高;X波段严重低估了树高,已经不适用这3种树高反演算法,反演的结果不理想。
Forest tree height is an important parameter of forest resources,so study on inversion of tree height of forest has important significance.This paper's data source has two full polarization data which were generated by PolSARPro software of ESA and two bands for the pretreatment,and the DEM difference method,RVOG phase method and the coherent amplitude method were chosen to reverse the tree height,finally analyzing and evaluating the result of L band and X band.The results show that for L band DEM difference method underestimates the tree height,RVOG phase method overestimates the tree height,the coherent amplitude inversion tree height precision is higher than DEM difference method and RVOG phase method,the X band badly underestimates the tree height and could not be applied to the three tree height inversion algorithm,the inversion result is not ideal.
[1] 李新武,郭华东.航天飞机极化干涉雷达数据反演地表植被参数[J].遥感学报,2002,6(6):424-429.
[2] 谈璐璐,杨立波,杨汝良.基于ESPRIT算法的极化干涉SAR植被高度反演研究[J].测绘学报,2011,40(3):296-300.
[3] 李文梅.极化干涉SAR层析估测森林垂直结构参数方法研究[D].北京:中国林业科学研究院,2013.
[4] Treuhaft R N,Cloude S R.The Structure of Oriented Vegetation from Polarimetric Interferometry[J].IEEE Transactions on Geoscience and Remote Sensing,1999,37(5):2620-2624.
[5] 李哲.极化干涉SAR平均树高反演[D].兰州:中国科学院寒区旱区环境与工程研究所,2009.
[6] 李哲,陈尔学,王建,等.几种极化干涉SAR森林平均高反演算法的比较评价[J].遥感技术与应用,2009,24(5):611-616.
[7] Cloude S R.Polarimetry:The characterisatton of polarisatton effectsd in em scattering[D].Birmingham:University of Birmingham,1986.
