基于InSAR属性分类的地面沉降精确监测研究

Research on Precise Insar Monitoring of Land Subsidence with Attribute Classification

针对常规合成孔径雷达干涉测量(Interferometric Synthetic Aperture Radar,InSAR)地面沉降监测精确性不足的问题,以北方某研究区为例,采用一种基于时序InSAR空间数据、地理信息系统(Geographic Information System,GIS)数据库的InSAR属性分类精确监测方法。首先采用常规时序InSAR处理得到大范围、高密度的地表沉降监测数据,然后基于InSAR监测点的高精度三维空间位置、具有丰富地物信息的GIS数据库进行多源数据的空间拓扑分析,实现地面InSAR监测点的精确提取,并结合水准数据进行属性分类前后的对比验证分析,可将同一实验数据的最大误差从7.77 mm提升至4.68 mm,将均方差从5.22 mm提升至2.91 mm,从而验证本文方法的有效性。

To solve the insufficient precision problem of conventional InSAR ground subsidence monitoring,this paper took a research area in the north as an example and adopted an accurate monitoring method of InSAR attribute classification based on time series InSAR spatial data and GIS database.First,conventional time-series InSAR processing was apllied to obtain large-scale,high-density land surface subsidence monitoring data,then,based on the high-precision three-dimensional spatial position of InSAR monitoring points and the GIS database with rich surface information,the spatial topology analysis of multi-source data was carried out to realize ground-based InSAR accurate extraction of monitoring points,combined with level data for comparative verification analysis before and after attribute classification,the maximum error of the same experimental data can be increased from 7.77 mm to 4.68 mm,and the mean square error can be increased from 5.22 mm to 2.91 mm,thus verifying the effectiveness of the method in this paper.