粗DEM辅助的广域PS-InSAR数据预处理方法及应用

DEM-Based Wide-Swath PS-InSAR Pre-Processing Method for Ground Deformation Surveillance and Its Application

PS-InSAR技术是一种高精度的地表形变探测方法,小区域PS-InSAR处理的研究较为深入,但处理方法并不适用广域的形变探测和数据处理方法。本文针对卫星实时轨道精度差的问题,首先提出利用辅助DEM进行斜距误差和轨道系统误差校正提升差分干涉相位反演精度的方法;其次,针对广域图像配准误差的空变性,提出了一种由粗到精的层级配准方法,解决了配准误差空变对PS-InSAR永久散射体选取和图像相干性的影响;最后,通过对形变观测结果的地理编码结合加权处理,实现对超广域观测图像的拼接。利用仿真数据验证了粗DEM对系统误差校正的有效性,并进一步通过Sentinel-1A的2000多景雷达影像对云南省全境39.4万平方千米的覆盖区域进行了处理,结果表明了本文方法对系统误差校正和对广域形变探测处理的有效性。

PS-InSAR is a powerful technology for high accurate ground deformation detection.The processing flowchart of PS-InSAR for small observed scenes is investigated thoroughly,but it cannot be directly applied to the super wide-swath scene deformation detection and SAR data processing.Herein,according to the problem of real time ephemeris with low accuracy,a systematic timing error and orbit error calibration method is proposed based on the auxiliary digital elevation model(DEM),which improves the accuracy of the terrain inteferogram simulation.Then,for the variation of SAR images’coregistration errors,the hierarchical“coarse to fine”coregistration method is proposed,which eliminates the effect of the variation of coregistration errors on permanent scatterer selection and the coherence degradation.Finally,the weighted combination method with geocoded images is presented for the combination of multiple deformation results.The DEM-based system error correction method is validated by using the simulated datasets.Furthermore,more than 2000 images of Sentinel-1A,co-vered 394 thousand square kilometers of Yunnan province,are used to investigate the effectiveness of the proposed method.The processing results show the validity of the proposed method for system error calibration and wide-swath deformation surveillance.