| Ground subsidence poses a serious hazard to transportation,urban construction and the natural environment,not only restricting sustainable social and economic development,but also threatening the safety of human life and property.Permanent Scatterer Synthetic Aperture Radar Interferometry(PS-InSAR)technology is widely used in ground deformation monitoring because of its all-day,all-weather,wide measurement range and high resolution.This paper focuses on the selection method of PS points in PS-InSAR technology,and analyzes and summarizes the problems in various PS point selection methods,including the problem that the reliability of PS points is reduced due to the noise phase in the interferogram,the problem that the density of PS points is low,and the problem that PS points are misjudged and missed in the single threshold method.To address these problems,this paper proposes a three-step identification and selection method for adaptive time-varying filtered PS points that takes into account phase characteristics.The selection of PS points and their application in ground settlement monitoring are investigated in some detail in the paper,and the main research contents and results are as follows:(1)In response to the problem of degraded reliability of PS points due to noisy phases in interferograms,an ESMD-Lee adaptive time-varying filtering method is proposed and implemented in this paper.The results from numerical simulation experiments and real data experiments show that the filtering method proposed in this paper can not only effectively reduce the influence of the noise phase,but also avoid the loss of phase information and interference fringes.After comparing with several different filtering methods,including circular period mean filtering,circular period median filtering,2D-WT filtering and Lee filtering,the evaluation indexes show that the RMSE value of the proposed ESMD-Lee filtering method is 0.547,the EPI value is 0.7148 and the NOR value is 614,which are more reliable than several other conventional filtering methods.The results are more reliable than those of other conventional filtering methods,and lay a foundation for the subsequent selection of high-quality PS points.(2)To address the problems of poor reliability and low density of PS points in the PS-InSAR technique in the PS point identification and selection method,as well as the problems of misclassification and omission in the selection process.This paper proposes and implements an adaptive time-varying filtered PS point preference method that takes into account the phase characteristics.The method firstly considers the high stability of PS point scattering and uses the amplitude deviation index threshold method to select as many PS points as possible to complete the coarse selection of PS points;secondly,based on the ESMD-Lee adaptive time-varying filtering method,each interferometric phase map is filtered to reduce the influence of the noise phase and obtain a pure phase interferogram,and uses the coherence coefficient threshold method to complete the selection of PS points from the filtered interferometric phase map.The selection of PS points is completed;finally,the characteristics of PS points themselves are taken into account from both the amplitude and phase aspects,and the coarse selection of PS points and the selected PS points are combined to achieve the optimal selection of PS points.Finally,the PS points are selected by combining the rough PS points and the selected PS points.The accuracy of the PS points is improved and accurate ground deformation information is obtained.(3)In this paper,the area around the junction of Daxing District in Beijing and Gu’an City in Hebei Province is selected as the experimental area,and the level measurement data in this experimental area is used as the standard.The PS point distribution map shows that the PS point selection method proposed in this paper can solve some of the problems of misselection and improve the quality of PS points.The measured level data verifies that the ground deformation information obtained by the PS point selection method proposed in this paper is more dependable and accurate. |
