Study On Weighted Superposition Method Of InSAR Interferogram For Land Subsidence

With the accelerating process of global urbanization,a large number of urban infrastructure has been built,underground resources have been overexploited,and urban land subsidence is becoming more and more serious.Land subsidence can cause serious damage to buildings and production facilities,is not conducive to urban construction and resource exploration and development,and causes significant harm to the safety of people's lives and property.Although the traditional deformation monitoring and leveling methods have high precision,they are only suitable for the measurement of points and lines in a small range,and cannot effectively monitor a large range of areas.In recent years,the Interferometric Synthetic Aperture Radar(InSAR)technology has become the focus of researchers,which has the characteristics of all-day time,large range,high precision and high resolution.It shows great application potential in ground subsidence monitoring,which makes it possible to monitor surface deformation more accurately under the condition of limited data resources.In this thesis,based on the traditional Differential Interferometric Synthetic Aperture Radar(D-InSAR)monitoring method,a weighted interferogram superposition method based on a small number of images for time series monitoring is proposed.The main work of this paper is as follows:(1)A public master image optimal selection function model is proposed.Based on the comprehensive correlation function model,the function model normalized the three influencing factors of vertical space baseline,time baseline and Doppler centroid frequency difference,and added the exponential factor to reduce the influence of the three parameters caused by the dimensional difference.Multiple main images are selected by this model to form more interference pairs with higher coherence in a limited data set,which provides sufficient data support for the subsequent weighted stacking process of interferograms.(2)The interferogram with low coherence generated by traditional single main image will reduce the monitoring accuracy when it participates in the process of interference stacking.To solve this problem,a weighted stacking method of InSAR interferogram based on multiple main images is proposed.This method uses fewer SAR images to generate interference atlas,and uses weighted stacking method to avoid interference images with low coherence,so as to realize high precision and long time deformation monitoring in a certain area.Firstly,multiple main images are selected to generate multiple interference atlas.Secondly,the phase information and the signal to noise ratio between the phase information and the atmospheric noise are significantly improved by superposition of the coherent graphs according to the weight.Finally,the average velocity map of surface deformation is obtained by transforming the phase information into deformation information.The proposed algorithm is verified by the Sentinel-1A satellite's SAR images covering San Diego County,California,USA from July 2019 to December 2019,and the Sentinel-2A optical remote sensing data and groundwater data of the United States Geological Survey are compared and analyzed.The experimental results show that: In this paper,the weighted stacking method of multi-main image interferogram can realize the monitoring of time series deformation in the study area by using fewer data sets,and the monitoring results have high accuracy,the method is effective and feasible.