Vinícius Rafael Neris dos Santos

Automatic classification of metallic targets using pattern recognition of GPR reflection: a study in the IAG-USP Test Site, Sao Paulo (Brazil)

In this work, a methodology to automatically classify of metal targets using pattern recognition techniques on GPR reflection data is presented. The methodology consists of designing a multilayer perceptron (MLP) classifier based on features extracted from the targets in the subsoil, and then using it to classify hyperbolas diffraction indicating their position and depth. The classification of reflections allows a high resolution reconstruction of the subsurface with reduced computing time. The system was developed in MATLAB and applied to GPR data obtained at IAG-USP test site, located in the city of Sao Paulo, Brazil, where metallic drums were studied under controlled field conditions. This site contains different targets of variable sizes buried under different depths and it served as a model for the computational experiment. The results indicate that the automatic classification of the metallic targets in the subsoil is efficient, contributing for the reduction of the ambiguities in the geophysical data interpretation, besides having application on the subsoil mapping of utilities.

Application of time-reversal-based processing techniques to enhance detection of GPR targets

In this paper we analyze the performance of time-reversal (TR) techniques in conjunction with various Ground Penetrating Radar (GPR) pre-processing methods aimed at improving detection of subsurface targets. TR techniques were first developed for ultrasound applications and, by exploiting the invariance of the wave equation under time reversal, can yield features such as superresolution and statistical stability. The TR method was examined here using both synthetic and actual GPR field data under four different pre-processing strategies on the raw data, namely: mean background removal, eigenvalue background removal, a sliding-window space-frequency technique, and a noise-robust spatial differentiator along the scan direction. Depending on the acquisition mode, it was possible to determine with good precision the position and depth of the studied targets as well as, in some cases, to differentiate the targets from nearby clutter such as localized geological anomalies. The proposed methodology has the potential to be useful when applied to GPR data for the detection of buried targets.

Joint Inversion of Apparent Conductivity and Magnetic Susceptibility to Characterize Buried Targets

A good knowledge of the underground space is needed in urban planning. The main objects found buried are water and sewage pipes, cables, and contaminants storage tanks. Through the inversion of inductive electromagnetic data, we can obtain parameters like depth, position, and radius of these targets. Here, we show a joint inversion approach of apparent conductivity and magnetic susceptibility using data collected at the IAG/USP test site. The model employed assumes three coils (transmitter, receiver, and target/conductor) and relates the electromotive force, through the mutual inductance that occurs between the pairs of transmitter–receiver and conductor–receiver coils, to the acquired conductivity and susceptibility data. The achieved adjustment between field and calculated data indicates that, despite the simplicity of the model, it is a good representation of the conductors, increasing the results of the inversion process. The inversion algorithm is based on the controlled random search algorithm, which presented to be quite stable with electromagnetic data. The main advantage of the proposed approach is the relatively fast inversion process that showed better results than the traditional qualitative interpretation usually applied to this kind of investigation, being promising for noninvasive target characterization and having application in geotechnical studies.