Towards underground void detection with in-pipe ground penetrating radar

H. Noshahri*, M. van Delft, L. Olde Scholtenhuis, J. Franco Hempenius, E. Dertien

*Corresponding author for this work

Research output: Contribution to conferencePaperpeer-review

2 Citations (Scopus)


Ground Penetrating Radar (GPR) is an electromagnetic inspection method that is widely used to help locate and assess conditions of sewer pipes. However, when operated from the ground surface, the method is not efficient in identifying early stages of void formation, which if untreated, can lead to the appearance of sinkholes. To remedy this issue, a few studies have introduced in-pipe GPR inspection systems to identify structural defects and the voids behind the sewer wall. Still, less research has focused on studying the implications of emission, propagation, and reception of electromagnetic waves in the enclosed environment of sewer pipes and their subsequent impact on the resulting radargrams. In this study, we address this matter by modeling and comparing equivalent planar (ground-surface) and cylindrical (in-pipe) operation environments. Our results indicate the differences between the radargrams of the cylindrical and planar topologies in terms of intensity, the slope of hyperbolas, and presence of horizontal reflection lines. These results encourage practical considerations for conducting in-pipe GPR surveys with respect to choosing antenna separation and target depth of the survey and tuning the parameters for hyperbolas in automatic object detection algorithms.

Original languageEnglish
Publication statusPublished - Nov 2020
Event3rd Asia Pacific Meeting on Near Surface Geoscience & Engineering, NSGE 2020 - Virtual Conference
Duration: 2 Nov 20205 Nov 2020
Conference number: 3


Conference3rd Asia Pacific Meeting on Near Surface Geoscience & Engineering, NSGE 2020
Abbreviated titleNSGE 2020
CityVirtual Conference


Dive into the research topics of 'Towards underground void detection with in-pipe ground penetrating radar'. Together they form a unique fingerprint.

Cite this