Buried Utiltiy Detection Using GPR


Research project by

Sushil Suvarna, Graduate Student, Computer Science (sssuvarn@unity.ncsu.edu)
Dr. L.E. Bernold, Associate Professor, Dept. of Civil Engg.
Dr.W.Snyder, Associate Professor, Dept. of Electrical and Computer Engg.

Overview/Background :

The demand for new buried utilities, such as gas- and power lines, fiber-optic communication lines is growing with new construction, re-construction, and the growth of the communication infrastructure worldwide. As a result, utility contractors are busy digging and trenching into the ground in order to bury new utilities. Because the machinery for placing the new utilities underground, such as backhoe excavators, trenchers, augers, drills, and plows, don't "feel" when they are getting close to already buried object, utilities are easily damaged. Despite great efforts in locating existing utilities before a contractor is allowed to dig, accidents occur in great numbers. A key problem of the presently available hand-held technologies to detect underground utilities before digging is the inability to reliably predict the exact location (e.g., depth, of existing utilities).

Ground Penetrating Radar :

Ground Penetrating Radar (GPR) is a short-range system for remote sensing, which measures short pulse electromagnetic (EM) reflections due to variations of the electrical properties of the investigated medium. GPR uses high frequency pulsed electromagnetic waves to acquire subsurface information. The electromagnetic wave, which is radiated from a transmitting antenna, travels through the material at a velocity that is relative to the electrical properties of the material. As the wave propagates, if it hits an object or a boundary with different electrical properties, then part of the wave energy is reflected or scattered back to the source and the rest keeps propagating. The wave that is reflected back is captured by an antenna and an image is created that is reflective of the materials and boundaries present beneath the surface. Radar penetration and resolution tend to be reduced by the EM reflections with a range that goes from a few meters in conductive media, to 50m at most for low conductivity materials. The detection of voids, pipes, buried objects and interfaces rely on the spatial transient given by the change in permittivity. The main drawback with a GPR is the absence of detecting the exact material of the object buried.

Research Objective and Laboratory Setup :
Laboratory Setup
Most GPR image processing algorithms are based on a signature database that maps the different possible objects with their orientations and the images created by these objects under different soil depths and conditions. The optimal algorithm processes the image of the unknown object and compares it with those present in the database and generates an approximate estimation about the nature of the object. To study the responses of different pipe materials and other miscellaneous objects that could be present underground an experimental workspace has been setup.
Most sample tests performed in the detection of mines or pipes, involved moving the GPR in one line, either forward or backwards. However, our setup aims at observing images in a single plane initially, but evolving such a technique into one that considers the perpendicular movement of the GPR as well. This would result in an overall zigzag movement, that would generate a 3 dimensional GPR image to study
Sample GPR image of a buried metallic rod
The setup shown above is being used to study patterns generated by pipes of different materials, kept at various depths and also patterns generated by objects such as rocks and wood that might be present. At present an attempt to generate an ideal response, i.e a response in the absence of miscellaneous objects and soil irregularities, which could help process an unknown image, is on. The responses from various materials have been collected and the images are being studied to locate patterns unique to each object. The next stages include studying patterns generated by the zigzag motion of the GPR and create a 3 dimensional image that could be easy to interpret. Once the pattern recognition for pipe like structures has been studied, the next stage would be mounting the GPR with the EMI and obtain a better estimation of the presence or absence of buried pipes.
Back