Nonlinear Inversion for Multiple Objects in Transient Electromagnetic Induction Sensing of Unexploded Ordnance: Technique and Applications

Song, Lin-Ping; Pasion, Leonard R.; Billings, Stephen; Oldenburg, Douglas W.

doi:10.1109/tgrs.2011.2132138

Cited by 22 publications

(21 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Song et al [7,8] have shown that the success of multi-target inversions is critically dependent on high-quality starting models. Song attacked this problem by Monte Carlo generation of potential starting models, and using those models whose simulated data reasonably fit the actual data.…”

Section: Reduction-to-pole and Downward Continuationmentioning

confidence: 99%

“…Several groups have been experimenting with multiple target solvers. Song et al [7,8] use a Monte Carlo method to find starting models with a reasonably good data misfit, and then polished the parameter estimates using conventional decent inversion techniques. Keiswetter [10] uses a Monte Carlo technique to generate an initial population of models, then selects the best combination from this population, and finally iterates this process as needed.…”

Section: Multi-target Inversionmentioning

confidence: 99%

See 1 more Smart Citation

Expanded Processing Techniques for EMI Systems

Prouty¹,

Snyder²,

George³

et al. 2012

View full text Add to dashboard Cite

Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number.

show abstract

Section: Reduction-to-pole and Downward Continuationmentioning

confidence: 99%

Section: Multi-target Inversionmentioning

confidence: 99%

Expanded Processing Techniques for EMI Systems

Prouty¹,

Snyder²,

George³

et al. 2012

View full text Add to dashboard Cite

show abstract

“…This point dipole model tends to work best for relatively small and distant targets, and in terms of predictive accuracy and consistency has been superseded by generalizations that distribute dipole-moment densities over surfaces or volumes (Shubitidze et al, 2010b(Shubitidze et al, , 2010c. On the other hand, all of these models fare less well when analyzing anomalies due to more than one target: The inversion becomes more computationally expensive, has more local minima, and may require regularization (Song et al, 2009(Song et al, , 2011Grzegorczyk et al, 2011). When the number of targets is not known a priori, as in all real-world tests, the situation is complicated further because the models must be run several times (assuming different target numbers) and have the tendency to overfit the data.…”

Section: Introductionmentioning

confidence: 99%

Joint diagonalization applied to the detection and discrimination of unexploded ordnance

et al. 2012

View full text Add to dashboard Cite

Efforts to discriminate buried unexploded ordnance from harmless surrounding clutter are often hampered by the uncertainty in the number of buried targets that produce a given detected signal. We present a technique that helps determine that number with no need for data inversion. The procedure is based on the joint diagonalization of a set of multistatic response (MSR) matrices measured at different time gates by a time-domain electromagnetic induction sensor. In particular, we consider the Naval Research Laboratory's Time-Domain Electromagnetic Multisensor Towed Array Detection System (TEMTADS), which consists of a 5 × 5 square grid of concentric transmitter/receiver pairs. The diagonalization process itself generalizes one of the standard procedures for extracting the eigenvalues of a single matrix; in terms of execution time, it is comparable to diagonalizing the matrices one by one. We present the method, discuss and illustrate its mathematical basis and physical meaning, and apply it to several actual measurements carried out with TEMTADS at a test stand and in the field at the former Camp Butner in North Carolina. We find that each target in a measurement is associated with a set of nonzero timedependent MSR eigenvalues (usually three), which enables estimation of the number of targets interrogated. These eigenvalues have a characteristic shape as a function of time that does not change with the location and orientation of the target relative to the sensor. We justify analytically and empirically that symmetric targets have pairs of eigenvalues with constant ratios between them.

show abstract

“…To accurately extract polarization tensor parameters from the EMI response, we need reasonably accurate estimates of the location of the source. We also need to decide if the observed anomaly is due to a single object or two or more objects (Bell, 2006;Grzegorczyk et al, 2009;Song et al, 2009;Miller et al, 2010;Shubitidze et al, 2010;Song et al, 2011). Recent advances in EMI instrumentation have produced systems that are able to acquire multistatic/multicomponent data.…”

Section: Introductionmentioning

confidence: 99%

“…When using electromagnetics for UXO problems, the standard processing method of finding buried sources is to formulate a nonlinear data-fitting optimization problem that is solved for source location and polarizations of the UXO from an initial model (Pasion and Oldenburg, 2001;Smith and Morrison, 2004;Bell, 2006;Grzegorczyk et al, 2009;Song et al, 2009;Shubitidze et al, 2010;Song et al, 2011). However, the problem of finding source location can be completely separated from the estimation of polarization spectra.…”

Section: Introductionmentioning

confidence: 99%

Estimating source locations of unexploded ordnance using the multiple signal classification algorithm

Song

Oldenburg

Pasion³

2012

GEOPHYSICS

Self Cite

View full text Add to dashboard Cite

To extract intrinsic polarization parameters of a buried object from electromagnetic induction (EMI) responses, one has to first find its location. We developed an efficient method to find the approximate locations of multiple ordnance items using time domain electromagnetic data. The procedure was based upon the principle of multiple signal classification which exploits the orthogonality of signal and noise subspaces of multistatic EMI data. For an arbitrary multistatic array, we formulated transmitter-based and receiver-based imaging or steering vectorial operators that related with the left and right singular vectors of a multistatic response matrix. The operators were computed at potential source locations and are mapped onto the noise subspaces derived from data. A spatial metric function was therefore introduced to measure the magnitude of the projection. A 3D source imaging of the metric function could be obtained by evaluating all potential locations over a region of interest. In ideal cases, the perfect orthogonality between the computed signal subspace and measured noisy subspace can be achieved at or near a target location, and rendered an imaging peak at that location. Conversely, the peak image locations obtained from this technique were used as the indicators for where targets were most likely present. The number of targets could be estimated from the rank of the data matrix, provided there were a sufficient number of transmitters and receivers. In some instances the locations of multiple targets were imaged directly, but the procedure was enhanced by stripping the effect of a larger or shallower target from the image. The technique was evaluated using the test-stand and field data, and compared with the standard nonlinear inversion. The results showed that it has potential capability to accurately localize sources in EMI sensing.

show abstract

Nonlinear Inversion for Multiple Objects in Transient Electromagnetic Induction Sensing of Unexploded Ordnance: Technique and Applications

Cited by 22 publications

References 43 publications

Expanded Processing Techniques for EMI Systems

Expanded Processing Techniques for EMI Systems

Joint diagonalization applied to the detection and discrimination of unexploded ordnance

Estimating source locations of unexploded ordnance using the multiple signal classification algorithm

Contact Info

Product

Resources

About