有偏迭代估计在短基线集形变模型中的应用

进行SBAS技术时,选择的干涉对一般会出现一景影像和其他几景影像干涉的现象,导致构建的线性形变模型病态。在这种情况下,最小二乘法求解SBAS线性形变模型已不太适合。针对这一问题,本文结合Liu估计的有偏迭代法在求解病态线性模型时能够克服病态性对结果的影响,提出将此方法应用到求解SBAS线性形变模型中。以覆盖济宁地区的13景ENVISAT ASAR数据展开了SBAS技术在城市地面沉降监测中的应用实验,分别利用LS法、基于Liu估计的有偏迭代法解算形变模型。结果表明,基于Liu估计的有偏迭代法获得研究区的年平均沉降速率标准差比LS法获得的低1.4,且其求解的结果相较于LS法获得的结果更准确、稳定。

Application of the biased iterative estimation in the SBAS InSAR deformation model

During using small baseline InSAR technique,the appearance that one image can interfere several other images generally occurs when interferogram pairs are generated,which will lead the linear deformation model ill-conditioned.In this case,the least square method is not suitable for solving the SBAS linear deformation model.Because of the biased iteration method based on Liu estimation can overcome the influence of ill-condition on the results of the ill-conditioned linear model,so this paper applied the method to solve SBAS linear deformation model.The paper selected 13scenes ENVISAT ASAR images which covered Jining area to monitor the trend of surface deformation using SBAS InSAR technique.By the LS method,the iterative estimation method based on the Liu estimation respectively to solve the SBAS deformation model,the results show that the standard deviation of the annual mean subsidence rate obtained by the biased iteration method is 1.4lower than that obtained by the LS method,and its solution is more accurate and stable than that obtained by the LS method.