Energy Analysis in Semiautomatic and Automatic Velocity Estimation for Ground Penetrating Radar Data in Urban Areas: Case Study in Ho Chi Minh City, Vietnam

Abstract

Maps of underground construction works, such as water pipes and water drainage systems are necessary for expansion of urban areas. For shallow depths, Ground Penetrating Radar (GPR) can provide high-resolution subsurface images. Electromagnetic velocity is crucial in time-to-depth conversion and imaging of the structures from the GPR section. Shielded common-offset antennas can work in city surroundings due to superior noise isolation properties. We have implemented new automatic/semiautomatic strategies to define the electromagnetic velocity and locations of the construction works by using common offset GPR data. In our approach, Kirchhoff migration is employed to image underground objects by correcting the locations of subsurface reflectors (i.e. diffractors, dips). The automatic technique helps define the velocity and position of an object or a diffractor by targeting high-valued data points in the maximum energy difference section, which is calculated from multiple migrated GPR sections of different velocities. When migrated correctly, a collapsed diffractor will contain the majority of its energy at the peak of the diffraction hyperbola. If migrated using the wrong velocity, the peak of the diffraction hyperbola will contain the least energy, with the rest of the energy smeared over migration artefacts. In the semiautomatic technique, the calculated velocities and positions from the first strategy can help interpreters in judging focused zones and under/over migration artefacts from different migrated GPR sections by using a limited velocity band. We applied the techniques to 2D/3D visualizations of underground pipes from one numerical model and a case study in Ho Chi Minh City, Vietnam.