Application of variational Bayesian learning time series-based InSAR in urban track settlement monitoring
To enhance the safety performance of urban rail transit,this paper proposed and implemented a novel urban track settlement monitoring scheme.The scheme applied a variational Bayesian learning algorithm for precise analysis of time-series interferometric synthetic aperture radar(InSAR)data.Firstly,by using permanent scatterer-based InSAR technology,a total of 79 high-quality X-band synthetic aperture radar(SAR)images of the railway tracks in Nanchang City from January 2016 to September 2023 were acquired.Secondly,a series of image processing techniques were employed,including refined filtering of SAR images through variational Bayesian learning algorithm,precise distinction between primary and secondary images for interferometric processing,and accurate extraction of interferometric phase maps.As a result,deformation information of urban track target points was obtained.Finally,the deformation data was encoded with geographic coordinates and input into a professional geospatial data analysis platform for intuitive data visualization and in-depth analysis,yielding deformation data in the vertical direction.The experimental validation reveals that the proposed scheme can accurately identify target points within 35 meters on both sides of the track,reflecting the distribution of urban track settlement rates in the study area and thus uncovering the dynamic evolution process of urban track settlement.
万方数据
维普
