scholarly journals Particle Center Supported Plane for Subsurface Target Classification based on Full Polarimetric Ground Penetrating Radar

2019 ◽  
Vol 11 (4) ◽  
pp. 405
Author(s):  
Xuan Feng ◽  
Haoqiu Zhou ◽  
Cai Liu ◽  
Yan Zhang ◽  
Wenjing Liang ◽  
...  
Keyword(s):  
Ground Penetrating Radar ◽  
Decomposition Methods ◽  
Support Vector ◽  
Classification Methods ◽  
Target Classification ◽  
Data Set ◽  
Particle Center ◽  
Ground Penetrating ◽  
New Criterion

The subsurface target classification of ground penetrating radar (GPR) is a popular topic in the field of geophysics. Among the existing classification methods, geometrical features and polarimetric attributes of targets are primarily used. As polarimetric attributes contain more information of targets, polarimetric decomposition methods, such as H-Alpha decomposition, have been developed for target classification of GPR in recent years. However, the classification template used in H-Alpha classification is preset depending on the experience of synthetic aperture radar (SAR); therefore, it may not be suitable for GPR. Moreover, many existing classification methods require excessive human operation, particularly when outliers exist in the sample (the data set containing the features of targets); therefore, they are not efficient or intelligent. We herein propose a new machine learning method based on sample centers, i.e., particle center supported plane (PCSP). The sample center is defined as the point with the smallest sum of distances from all points in the same sample, which is considered as a better representation of the sample without significant effect of the outliers. In this proposed method, particle swarm optimization (PSO) is performed to obtain the sample centers; the new criterion for subsurface target classification is achieved. We applied this algorithm to full polarimetric GPR data measured in the laboratory and outdoors. The results indicate that, comparing with support vector machine (SVM) and classical H-Alpha classification, this new method is more efficient and the accuracy is relatively high.

scholarly journals Feature Extraction and Automatic Material Classification of Underground Objects from Ground Penetrating Radar Data

2014 ◽  
Vol 2014 ◽  
pp. 1-10 ◽  
Author(s):  
Qingqing Lu ◽  
Jiexin Pu ◽  
Zhonghua Liu
Keyword(s):  
Feature Extraction ◽  
Ground Penetrating Radar ◽  
Classification Accuracy ◽  
Support Vector ◽  
Discrete Wavelet ◽  
Noisy Environment ◽  
Feature Based ◽  
Approximation Coefficients ◽  
Ground Penetrating

Ground penetrating radar (GPR) is a powerful tool for detecting objects buried underground. However, the interpretation of the acquired signals remains a challenging task since an experienced user is required to manage the entire operation. Particularly difficult is the classification of the material type of underground objects in noisy environment. This paper proposes a new feature extraction method. First, discrete wavelet transform (DWT) transforms A-Scan data and approximation coefficients are extracted. Then, fractional Fourier transform (FRFT) is used to transform approximation coefficients into fractional domain and we extract features. The features are supplied to the support vector machine (SVM) classifiers to automatically identify underground objects material. Experiment results show that the proposed feature-based SVM system has good performances in classification accuracy compared to statistical and frequency domain feature-based SVM system in noisy environment and the classification accuracy of features proposed in this paper has little relationship with the SVM models.

Fast-HPLC Fingerprinting to Discriminate Olive Oil from Other Edible Vegetable Oils by Multivariate Classification Methods

2017 ◽  
Vol 100 (2) ◽  
pp. 345-350 ◽  
Author(s):  
Ana M Jiménez-Carvelo ◽  
Antonio González-Casado ◽  
Estefanía Pérez-Castaño ◽  
Luis Cuadros-Rodríguez
Keyword(s):  
Olive Oil ◽  
Vegetable Oils ◽  
Edible Oils ◽  
Support Vector ◽  
Classification Methods ◽  
Multivariate Classification ◽  
Chromatographic Fingerprint ◽  
K Nearest Neighbors ◽  
Pls Discriminant Analysis

Abstract A new analytical method for the differentiation of olive oil from other vegetable oils using reversed-phaseLC and applying chemometric techniques was developed. A 3 cm short column was used to obtain the chromatographic fingerprint of the methyl-transesterified fraction of each vegetable oil. The chromatographic analysis tookonly 4 min. The multivariate classification methods used were k-nearest neighbors, partial least-squares (PLS) discriminant analysis, one-class PLS, support vector machine classification, and soft independent modeling of class analogies. The discrimination of olive oil from other vegetable edible oils was evaluated by several classification quality metrics. Several strategies for the classification of the olive oil wereused: one input-class, two input-class, and pseudo two input-class.

A Comparison of Semi-Supervised Classification Approaches for Software Defect Prediction

2014 ◽  
Vol 23 (1) ◽  
pp. 75-82 ◽  
Author(s):  
Cagatay Catal
Keyword(s):  
Supervised Classification ◽  
Defect Prediction ◽  
Support Vector ◽  
Software Defect Prediction ◽  
Classification Methods ◽  
Data Set ◽  
Software Defect ◽  
Data Points ◽  
Supervised Classification Methods ◽  
Prediction Approach

AbstractPredicting the defect-prone modules when the previous defect labels of modules are limited is a challenging problem encountered in the software industry. Supervised classification approaches cannot build high-performance prediction models with few defect data, leading to the need for new methods, techniques, and tools. One solution is to combine labeled data points with unlabeled data points during learning phase. Semi-supervised classification methods use not only labeled data points but also unlabeled ones to improve the generalization capability. In this study, we evaluated four semi-supervised classification methods for semi-supervised defect prediction. Low-density separation (LDS), support vector machine (SVM), expectation-maximization (EM-SEMI), and class mass normalization (CMN) methods have been investigated on NASA data sets, which are CM1, KC1, KC2, and PC1. Experimental results showed that SVM and LDS algorithms outperform CMN and EM-SEMI algorithms. In addition, LDS algorithm performs much better than SVM when the data set is large. In this study, the LDS-based prediction approach is suggested for software defect prediction when there are limited fault data.

Helicopter-borne ground-penetrating radar investigations on temperate alpine glaciers: A comparison of different systems and their abilities for bedrock mapping

Geophysics ◽  
2016 ◽  
Vol 81 (1) ◽  
pp. WA119-WA129 ◽  
Author(s):  
Anja Rutishauser ◽  
Hansruedi Maurer ◽  
Andreas Bauder
Keyword(s):  
Ground Penetrating Radar ◽  
Low Frequency ◽  
Large Data ◽  
Internal Reflection ◽  
Swiss Alps ◽  
Reflection Band ◽  
Data Set ◽  
Investigation Area ◽  
Attractive Option ◽  
Ground Penetrating

On the basis of a large data set, comprising approximately 1200 km of profile lines acquired with different helicopter-borne ground-penetrating radar (GPR) systems over temperate glaciers in the western Swiss Alps, we have analyzed the possibilities and limitations of using helicopter-borne GPR surveying to map the ice-bedrock interface. We have considered data from three different acquisition systems including (1) a low-frequency pulsed system hanging below the helicopter (BGR), (2) a stepped frequency system hanging below the helicopter (Radar Systemtechnik GmbH [RST]), and (3) a commercial system mounted directly on the helicopter skids (Geophysical Survey Systems Incorporated [GSSI]). The systems showed considerable differences in their performance. The best results were achieved with the BGR system. On average, the RST and GSSI systems yielded comparable results, but we observed significant site-specific differences. A comparison with ground-based GPR data found that the quality of helicopter-borne data is inferior, but the compelling advantages of airborne surveying still make helicopter-borne data acquisition an attractive option. Statistical analyses concerning the bedrock detectability revealed not only large differences between the different acquisition systems but also between different regions within our investigation area. The percentage of bedrock reflections identified (with respect to the overall profile length within a particular region) varied from 11.7% to 68.9%. Obvious factors for missing the bedrock reflections included large bedrock depths and steeply dipping bedrock interfaces, but we also observed that internal features within the ice body may obscure bedrock reflections. In particular, we identified a conspicuous “internal reflection band” in many profiles acquired with the GSSI system. We attribute this feature to abrupt changes of the water content within the ice, but more research is required for a better understanding of the nature of this internal reflection band.

scholarly journals Underground Object Classification for Urban Roads Using Instantaneous Phase Analysis of Ground-Penetrating Radar (GPR) Data

2018 ◽  
Vol 10 (9) ◽  
pp. 1417 ◽  
Author(s):  
Byeongjin Park ◽  
Jeongguk Kim ◽  
Jaesun Lee ◽  
Man-Sung Kang ◽  
Yun-Kyu An
Keyword(s):  
Phase Analysis ◽  
Ground Penetrating Radar ◽  
Data Interpretation ◽  
Buried Pipes ◽  
Instantaneous Phase ◽  
Analysis Technique ◽  
Different Types ◽  
Ground Penetrating ◽  
Rapid Scanning

Ground-penetrating radar (GPR) has been widely used to detect subsurface objects, such as hidden cavities, buried pipes, and manholes, owing to its noncontact sensing, rapid scanning, and deeply penetrating remote-sensing capabilities. Currently, GPR data interpretation depends heavily on the experience of well-trained experts because different types of underground objects often generate similar GPR reflection features. Moreover, reflection visualizations that were obtained from field GPR data for urban roads are often weak and noisy. This study proposes a novel instantaneous phase analysis technique to address these issues. The proposed technique aims to enhance the visibility of underground objects and provide objective criteria for GPR data interpretation so that the objects can be automatically classified without expert intervention. The feasibility of the proposed technique is validated both numerically and experimentally. The field test utilizes rarely available GPR data for urban roads in Seoul, South Korea and demonstrates that the technique allows for successful visualization and classification of three different types of underground objects.

Identification, investigation and classification of surface depressions and chalk dissolution features using integrated LiDAR and geophysical methods

2020 ◽  
Vol 53 (4) ◽  
pp. 620-644 ◽  
Author(s):  
Zoe Elizabeth Jeffery ◽  
Stephen Penn ◽  
David Peter Giles ◽  
Linley Hastewell
Keyword(s):  
Ground Penetrating Radar ◽  
Image Interpretation ◽  
Geophysical Methods ◽  
Integrated Approach ◽  
Study Data ◽  
Aerial Photographs ◽  
Geophysical Techniques ◽  
Karstic Features ◽  
Ground Penetrating

The chalk bedrock of the Hampshire Basin, southern England is an important aquifer and is highly susceptible to dissolution, making the development and presence of karstic features a widespread occurrence. These features are hazardous because they provide possible pathways to the underlying aquifer and therefore present potential site-specific contamination risks. There is also evidence of extensive extraction, through both mining and surface quarrying, of chalk, flint and clay over many centuries. Geophysical techniques consisting of electromagnetic (EM31) and ground-penetrating radar surveys were used to identify and characterize target features identified from desk study data. The ground-penetrating radar and EM31 interpretations allowed the classification of non-anthropogenic target features, such as diffuse buried sinkholes with disturbed and subsiding clay-rich infill and varying symmetrical and asymmetrical morphologies. We describe here the investigations of such features identified at Holme Farm, Stansted House, Hampshire. The combination of EM31 data and ground-penetrating radar profiles facilitated the identification of a palaeovalley, cavities and irregular rockhead. This investigation identified locations of aquifer contamination risk as some sinkholes have been sites for the illegal dumping of waste or the infiltration of fertilizers, leaking sewage pipes or animal waste. This potential source of contamination utilizes the sinkhole as a pathway into the highly transmissive White Chalk Subgroup of Hampshire and has caused contamination of the aquifer. We conclude that our integrated approach of geophysical techniques linked to aerial photographs and LiDAR image interpretation was highly effective in the location and characterization of dissolution structures, infilled former quarries and mining features at this site.

On the use of dispersive APVO GPR curves for thin-bed properties estimation: Theory and application to fracture characterization

Geophysics ◽  
2009 ◽  
Vol 74 (1) ◽  
pp. J1-J12 ◽  
Author(s):  
Jacques Deparis ◽  
Stéphane Garambois
Keyword(s):  
Ground Penetrating Radar ◽  
Inverse Method ◽  
Phase Variation ◽  
Fractured Media ◽  
Data Set ◽  
Fracture Characterization ◽  
Propagation Effects ◽  
Phase Liquid ◽  
Ground Penetrating ◽  
Homogeneous Formation

The presence of a thin layer embedded in any formation creates complex reflection patterns caused by interferences within the thin bed. The generated reflectivity amplitude variations with offset have been increasingly used in seismic interpretation and more recently tested on ground-penetrating radar (GPR) data to characterize nonaqueous-phase liquid contaminants. Phase and frequency sensitivities of the reflected signals are generally not used, although they contain useful information. The present study aims to evaluate the potential of these combined properties to characterize a thin bed using GPR data acquired along a common-midpoint (CMP) survey, carried out to assess velocity variations in the ground. It has been restricted to the simple case of a thin bed embedded within a homogeneous formation, a situation often encountered in fractured media. Dispersive properties ofthe dielectric permittivity of investigated materials (homogeneous formation, thin bed) are described using a Jonscher parameterization, which permitted study of the dependency of amplitude and phase variation with offset (APVO) curves on frequency and thin-bed properties (filling nature, aperture). In the second part, we discuss and illustrate the validity of the thin-bed approximation as well as simplify assumptions and make necessary careful corrections to convert raw CMP data into dispersive APVO curves. Two different strategies are discussed to correct the data from propagation effects: a classical normal-moveout approach and an inverse method. Finally, the proposed methodology is applied to a CMP GPR data set acquired along a vertical cliff. It allowed us to extract the characteristics of a subvertical fracture with satisfying resolution and confidence. The study motivates interest to use dispersion dependency of the reflection coefficient variations for thin-bed characterization.

Sign in / Sign up

Export Citation Format

Share Document