变分贝叶斯学习时序InSAR在城市轨道沉降监测中的应用  

Application of variational Bayesian learning time series-based InSAR in urban track settlement monitoring

作　　者：刘明 LIU Ming(Geographic Information Engineering Brigade of Jiangxi Bureau of Geology,Nanchang,Jiangxi 330001,China)

机构地区：[1]江西省地质局地理信息工程大队,江西南昌330001

出　　处：《北京测绘》2024年第12期1771-1776,共6页Beijing Surveying and Mapping

摘　　要：为提升城市轨道交通的安全性能,本文提出并实现了一种新的城市轨道沉降监测方案,该方案的关键在于采用变分贝叶斯学习算法对时序合成孔径雷达干涉测量(InSAR)数据进行精准解析。首先,利用永久散射体InSAR技术,获取南昌市铁路线路轨道2016年1月至2023年9月总计79幅高质量的X波段合成孔径雷达(SAR)图像;其次,采用一系列图像处理手段,包括通过变分贝叶斯学习算法对SAR图像进行精细化滤波,精确区分主辅影像并进行干涉处理,精准提取干涉相位图,从而得到了城市轨道目标点形变信息;最后,对形变数据进行地理坐标编码,并输入到专业的地理空间数据分析平台,以便进行直观的数据可视化和深入的分析,最终获取了垂直方向上的形变数据。结果表明,本文方案能准确识别轨道及其两侧35 m范围内的目标点,反映研究区域内城市轨道的沉降速率分布状况,从而揭示了城市轨道沉降的动态演变过程。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 timeseries 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.

关 键 词：时序合成孔径雷达干涉测量数据 变分贝叶斯 城市轨道 沉降监测 

分 类 号：P258[天文地球—测绘科学与技术]