Monitoring Deformation along Railway Systems Combining Multi-Temporal InSAR and LiDAR Data

Fengming Hu,Jicang Wu,Ramon F Hanssen,Ling Chang,Freek J Van Leijen

doi:10.3390/rs11192298

Abstract

Multi-temporal interferometric synthetic aperture radar (MT-InSAR) can be applied to monitor the structural health of infrastructure such as railways, bridges, and highways. However, for the successful interpretation of the observed deformation within a structure, or between structures, it is imperative to associate a radar scatterer unambiguously with an actual physical object. Unfortunately, the limited positioning accuracy of the radar scatterers hampers this attribution, which limits the applicability of MT-InSAR. In this study, we propose an approach for health monitoring of railway system combining MT-InSAR and LiDAR (laser scanning) data. An amplitude-augmented interferometric processing approach is applied to extract continuously coherent scatterers (CCS) and temporary coherent scatterers (TCS), and estimate the parameters of interest. Based on the 3D confidence ellipsoid and a decorrelation transformation, all radar scatterers are linked to points in the point cloud and their coordinates are corrected as well. Additionally, several quality metrics defined using both the covariance matrix and the radar geometry are introduced to evaluate the results. Experimental results show that most radar scatterers match well with laser points and that LiDAR data are valuable as auxiliary data to classify the radar scatterers.

Highlights

Satellite-based differential interferometric synthetic aperture radar (DInSAR) is a standard geodetic technology for deformation monitoring over wide areas with millimeter accuracy [1]
A thermal dilation parameter is introduced to describe the variations of interferometric phase with temperature since thermal dilation often happens along the railway due to its steel structure [28,30]
Since the estimated heights of all scatterers are related to a specified reference, we propose a solution search method to estimate the height offset with the help of grid data obtained by LiDAR point cloud

Summary

Introduction

Satellite-based differential interferometric synthetic aperture radar (DInSAR) is a standard geodetic technology for deformation monitoring over wide areas with millimeter accuracy [1]. Railway systems consist of a complex collection of constructions, such as embankments, tunnels and bridges, subject to changing environmental conditions (geology, relief). Examples are the differential subsidence of assets in soft soils, slope instabilities/slow landslides in mountainous areas, embankment instabilities, and aging and degradation of concrete constructions. Current approaches for structural health monitoring are levelling, linear variable differential transformers and video based systems [14]. While the latter can be used to monitor dynamic displacements [15,16], their applicability is limited due to manual operation and localized implementation. MT-InSAR is complementary to these in situ techniques and has the advantage of wide area applications, frequent revisits, and a millimeter level precision

Methods

Results

Conclusion