Automated data extraction from synthetic and real radargrams of district heating pipelines

The main goal of this paper is to investigate the performance of an algorithm for point extraction from hyperbolic reflections in synthetic and real Ground-Penetrating Radar (GPR) data. The real radargrams that we considered contain hyperbolic reflections due to the presence, in the surveyed area, of district heating pipelines DN250 (250mm inner diameter pipe). These are buried 88 cm deep in a soil trench, and covered by compacted sand and concrete bricks (behaton pavement). The synthetic radargrams result from the simulation of a model representing the real geometry on the location of interest. The simulation was carried out by using gprMax, ver. 3.

[1]  Craig Warren,et al.  gprMax: Open source software to simulate electromagnetic wave propagation for Ground Penetrating Radar , 2016, Comput. Phys. Commun..

[2]  S. Fontul,et al.  COST Action TU1208 - Working Group 2 - GPR surveying of pavements, bridges, tunnels and buildings; underground utility and void sensing , 2017 .

[3]  Sven Birkenfeld Automatic detection of reflexion hyperbolas in gpr data with neural networks , 2010, 2010 World Automation Congress.

[4]  Christina Plati,et al.  Inspection Procedures for Effective GPR Sensing and Mapping of Underground Utilities and Voids, with a Focus to Urban Areas , 2015 .

[5]  Raffaele Persico,et al.  Automated Detection of Reflection Hyperbolas in Complex GPR Images With No A Priori Knowledge on the Medium , 2016, IEEE Transactions on Geoscience and Remote Sensing.

[6]  Michael Felsberg,et al.  Enhanced analysis of thermographic images for monitoring of district heat pipe networks , 2016, Pattern Recognit. Lett..

[7]  María de la Vega Pérez Gracia,et al.  Civil Engineering Applications of Ground Penetrating Radar , 2015 .

[8]  Chiara Delmastro,et al.  Underground urbanism: Master Plans and Sectorial Plans , 2016 .

[9]  Lars Schmidt-Thieme,et al.  Buried pipe localization using an iterative geometric clustering on GPR data , 2014, Artificial Intelligence Review.

[10]  Miro Govedarica,et al.  Integration of modern remote sensing technologies for faster utility mapping and data extraction , 2015 .

[11]  Nicole Metje,et al.  Underground asset location and condition assessment technologies , 2007 .