A combined strategy for landmine detection and identification using synthetic GPR responses

González-Huici, María A.; Giovanneschi, Fabio

doi:10.1016/j.jappgeo.2013.08.006

Cited by 28 publications

(17 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We evaluated the proposed approach with the measurement data from a 2013 field campaign at Leibniz Institute for Applied Geophysics (LIAG) in Hannover (Germany) [11]; Fig. 2 shows the test field, for detailed ground truth informations.…”

Section: B Test Field Measurementsmentioning

confidence: 99%

See 1 more Smart Citation

Dictionary Learning for Adaptive GPR Landmine Classification

Giovanneschi

Mishra

González-Huici

et al. 2019

IEEE Trans. Geosci. Remote Sensing

Self Cite

View full text Add to dashboard Cite

Ground penetrating radar (GPR) target detection and classification is a challenging task. Here, we consider various online dictionary learning (DL) methods to obtain sparse representation (SR) of the GPR data to enhance feature extraction for target classification via support vector machine. Online methods are preferred because traditional batch DL like K-SVD is not scalable to high-dimensional training sets and infeasible for real-time operation. We also develop Drop-Off MINibatch Online Dictionary Learning (DOMINODL) which exploits the fact that a lot of the training data may be correlated. The DOMINODL algorithm iteratively considers elements of the training set in small batches and drops off samples which become less relevant. For the case of abandoned anti-personnel landmines classification, we compare the performance of K-SVD with three online algorithms: classical Online Dictionary Learning, its correlation-based variant and DOMINODL. Our experiments with real data from L-band GPR show that online DL methods reduce learning time by 36-93% and increase mine detection by 4-28% over K-SVD. Our DOMINODL is the fastest and retains similar classification performance as the other two online DL approaches. We use a Kolmogorov-Smirnoff test distance and the Dvoretzky-Kiefer-Wolfowitz inequality for the selection of DL input parameters leading to enhanced classification results. To further compare with state-of-the art classification approaches, we evaluate a convolutional neural network (CNN) classifier which performs worse than the proposed approach. Moreover, when the acquired samples are randomly reduced by 25%, 50% and 75%, sparse decomposition based classification with DL remains robust while the CNN accuracy is drastically compromised.2 Furthermore, online DL 2 methods have been studied more generally in GPR. Only one other previous study has employed DL (K-SVD) using GPR signals [24], although for the application of identifying bedrock features. We employ online DL methods and use the coefficients of the resulting sparse vectors as input to a SVM classifier to distinguish mines from clutter. Our comparison of K-SVD and online DL using real data from L-band GPR shows that online DL algorithms present distinct advantages in speed and low false-alarm rates.2) A new, faster online DL algorithm. We propose a new Drop-Off MINi-batch Online Dictionary Learning (DOMINODL) which processes the training data in mini-batches and avoids unnecessary update of the irrelevant atoms in order to reduce the computational complexity. The intuition for the dropoff step comes from the fact that some training samples are uncorrelated and, therefore, in the interest of processing time, they can be dropped during training without significantly affecting performance.3) Better statistical metrics for improved classification. Contrary to previous studies [24] which determine DL parameters (number of iterations, atoms, etc.) based on bulk statistics such as normalized root-mean-square-error (NRMSE), we consider statistical inference for...

show abstract

Section: B Test Field Measurementsmentioning

confidence: 99%

“…A variety of signal processing algorithms have been proposed for detection of low metal-content landmines in realistic scenarios; approaches based on feature extraction and classification are found to be the most effective (see e.g. [10][11][12][13]), yet false-alarm rates remain very high.…”

Section: Introductionmentioning

confidence: 99%

Dictionary Learning for Adaptive GPR Landmine Classification

Giovanneschi

Mishra

González-Huici

et al. 2019

IEEE Trans. Geosci. Remote Sensing

Self Cite

View full text Add to dashboard Cite

show abstract

“…Commonly used techniques in landmine detection include correlation functions [23], cross-correlation with simulated samples [24] and least mean squares (LMS) methods or their variants, 2D LMS or 3D LMS [25]. Many researchers have begun to use statistical approaches [26], AdaBoost classifiers [27], hidden Markov models [4] and other techniques for landmine detection.…”

Section: Automated Detection Toolmentioning

confidence: 99%

GPR Signal Characterization for Automated Landmine and UXO Detection Based on Machine Learning Techniques

et al. 2014

View full text Add to dashboard Cite

Abstract:Landmine clearance is an ongoing problem that currently affects millions of people around the world. This study evaluates the effectiveness of ground penetrating radar (GPR) in demining and unexploded ordnance detection using 2.3-GHz and 1-GHz high-frequency antennas. An automated detection tool based on machine learning techniques is also presented with the aim of automatically detecting underground explosive artifacts. A GPR survey was conducted on a designed scenario that included the most commonly buried items in historic battle fields, such as mines, projectiles and mortar grenades. The buried targets were identified using both frequencies, although the higher vertical resolution provided by the 2.3-GHz antenna allowed for better recognition of the reflection patterns. The targets were also detected automatically using machine learning techniques. Neural networks and logistic regression algorithms were shown to be able to discriminate between potential targets and clutter. The neural network had the most success, with accuracies ranging from 89% to 92% for the 1-GHz and 2.3-GHz antennas, respectively.

show abstract

“…During the last few decades, ground‐penetrating radar (GPR) has received considerable interest when searching for buried objects (see, e.g., Boonpoonga, ; Cedrina et al, ; Gonzalez‐Huici & Giovanneschi, ; Wang et al, ; Xie et al, ) because of the ability to nondestructively detect buried objects beneath shallow earth surface. The GPR that can image the subsurface with a high spatial resolution offers the promise of detecting metallic and nonmetallic land mines.…”

Section: Introductionmentioning

confidence: 99%

Automatic Detection and Classification of Buried Objects Using Ground‐Penetrating Radar for Counter‐Improvised Explosive Devices

et al. 2018

View full text Add to dashboard Cite

In this paper, a technique that can automatically detect and classify objects buried under the ground is proposed. The technique employs a ground‐penetrating radar that transmits electromagnetic waves in order to strike the objects and then receives the backscattering electromagnetic wave to perform signal processing. This signal processing is divided into four main steps as follows. First, preprocessing is used to reduce the clutter due to the effect of the media layer interface. Second, the late time of the scattering signal is estimated using a simple cross correlation. Third, a few successive poles are extracted from the scattering response at the estimated late time by using the short‐time matrix pencil method. Finally, the extracted poles are fed for object classification with different constitutions and/or shapes using a support vector machine. Simulations according to the practical situation in three southern provinces of Thailand to counter the improvised explosive devices were set up. The performance of the proposed technique was evaluated. The simulation results showed that the proposed technique can efficiently detect and classify buried objects for counter‐improvised explosive device operations in the military.

show abstract

A combined strategy for landmine detection and identification using synthetic GPR responses

Cited by 28 publications

References 11 publications

Dictionary Learning for Adaptive GPR Landmine Classification

Dictionary Learning for Adaptive GPR Landmine Classification

GPR Signal Characterization for Automated Landmine and UXO Detection Based on Machine Learning Techniques

Automatic Detection and Classification of Buried Objects Using Ground‐Penetrating Radar for Counter‐Improvised Explosive Devices

Contact Info

Product

Resources

About