Time Delay and Permittivity Estimation by Ground-Penetrating Radar With Support Vector Regression

Bastard, Cédric Le; Wang, Yide; Baltazart, Vincent; Dérobert, Xavier

doi:10.1109/lgrs.2013.2280500

Cited by 44 publications

(25 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The problem of detecting buried targets through electromagnetic waves is a classical inverse scattering problem which is relevant in a number of different applications [1,2] such as civil engineering diagnostics [3][4][5][6][7][8][9], archaeological and geophysical prospecting [10,11], cultural heritage monitoring [12,13], mines and IEDs (improvised explosive devices) detection [14], only to mention a few of them. Electromagnetic waves at the microwave spectrum allow to get high resolution.…”

Section: Introductionmentioning

confidence: 99%

Subsurface Detection of Shallow Targets by Undersampled Multifrequency Data and a Non-Cooperative Source

et al. 2019

View full text Add to dashboard Cite

Imaging buried objects embedded within electrically large investigation domains can require a large number of measurement points. This is impractical if long data acquisition time cannot be tolerated or the system is conceived to work at some stand-off distance from the air/soil interface; for example, if it is mounted over some flying platform. In order to reduce the number of spatial measurements, here, we propose a method for detecting and localizing shallowly buried scattering targets from under-sampled far-field data. The method is based on a scattering model derived from the equivalence theorem for electromagnetic radiation. It exploits multi-frequency data and does not require that the transmitter and receivers are synchronized, making the source non-cooperative. To provide a benchmark against which spatial data have to be reduced, first, the number of required spatial measurements is examined by analyzing the properties of the relevant scattering operator. Then, since under-sampling data produces aliasing artifacts, frequency diversity (i.e., multi-frequency data) is exploited to mitigate those artifacts. In particular, single-frequency reconstructions are properly fused and a criterion for selecting the frequencies to be used is provided. Numerical examples show that the method allows for satisfactory target transverse localization with a number of measurements that are much less than the ones required by other methods commonly used in subsurface imaging.

show abstract

Section: Introductionmentioning

confidence: 99%

Subsurface Detection of Shallow Targets by Undersampled Multifrequency Data and a Non-Cooperative Source

et al. 2019

View full text Add to dashboard Cite

show abstract

“…The feature vector is then extracted from the correlation matrix [24]. Since the correlation matrix is Hermitian, only the upper triangular half of the matrix is used, and the elements in there are reorganized to form the feature vector v as follows:…”

Section: The Regular Svr Methods In Wpementioning

confidence: 99%

“…Originally, SVM is introduced to identify and classify the targets from radar images and high-resolution range profiles [15][16][17], and then applied into more applications, e.g., ground-penetrating radar (GPR) and ocean clutter suppression [18][19][20]. Recently, SVR has also been involved in radar areas to solve regression problems, such as prediction of the vehicle travel time [21] and parameter estimation in GPR [22][23][24]. Inspired by these successful applications, this paper brings SVR into TWRS to estimate the wall parameters and uses it in the regular way, which can be noted as the regular SVR method.…”

Section: Introductionmentioning

confidence: 99%

Wall parameters estimation based onsupport vector regression for through wall radar sensing

Chen

2015

EURASIP J. Adv. Signal Process.

View full text Add to dashboard Cite

In through wall radar sensing, the wall parameters estimation (WPE) problem has been a topic that attracts a lot of attention since the wall parameters, i.e., the permittivity and the thickness, are of crucial importance to locate the targets and to produce a well-focused image, but they are usually unknown in practice. To solve this problem, in this paper, the support vector regression (SVR), a powerful tool for regression analysis, is introduced, and its performance on WPE, provided it is used it in the regular way, is investigated. Unfortunately, it is shown that the regular use of SVR cannot afford satisfactory estimation results since the sample data used in SVR, namely the received echoes from the walls, are seriously interfered with the echoes from the targets which are located near the walls. In view of this limitation, a novel SVR-based WPE approach that consists of three stages is proposed by this paper. In the first stage, three regression functions are trained by SVR, one of which will output the estimate of the permittivity in the second stage, and the others are designed to output two instrumental variables for estimating the thickness. In the third stage, the estimate of thickness will be achieved by minimizing a predefined cost function wherein the estimated permittivity and the outputted instrumental variables are involved. The better robustness and higher estimation accuracy of the proposed approach compared to the regular use of SVR are validated by the numerical experimental results using finite-difference time-domain simulations.

show abstract

“…For example, in [62] an efficient algorithm for the computation of the time domain reflection coefficient in the transverse magnetic (TM) case has been developed in order to better characterize the surface reflections. In [63], [64] subspace methods and learning-based strategies have been used for the estimation of time delays, permittivity and roughness parameters within pavement structures. Estimation of specific soil parameters (e.g., moisture and clay content) have also been addressed in [65].…”

Section: Soil Properties Estimationmentioning

confidence: 99%

Advanced Inversion Techniques for Ground Penetrating Radar

Fedeli¹,

Pastorino²,

Randazzo³

2017

JTIT

View full text Add to dashboard Cite

Abstract-Ground Penetrating Radar (GPR) systems are nowadays standard inspection tools in several application areas, such as subsurface prospecting, civil engineering and cultural heritage monitoring. Usually, the raw output of GPR is provided as a B-scan, which has to be further processed in order to extract the needed information about the inspected scene. In this framework, inverse-scattering-based approaches are gaining an ever-increasing interest, thanks to their capabilities of directly providing images of the physical and dielectric properties of the investigated areas. In this paper, some advances in the development of such inversion techniques in the GPR field are revised and discussed.

show abstract

Time Delay and Permittivity Estimation by Ground-Penetrating Radar With Support Vector Regression

Cited by 44 publications

References 28 publications

Subsurface Detection of Shallow Targets by Undersampled Multifrequency Data and a Non-Cooperative Source

Subsurface Detection of Shallow Targets by Undersampled Multifrequency Data and a Non-Cooperative Source

Wall parameters estimation based onsupport vector regression for through wall radar sensing

Advanced Inversion Techniques for Ground Penetrating Radar

Contact Info

Product

Resources

About