Deep learning-based automated underground cavity detection using three-dimensional ground penetrating radar

Three-dimensional ground penetrating radar data are often ambiguous and complex to interpret when attempting to detect only underground cavities because ground penetrating radar reflections from various underground objects can appear like those from cavities. In this study, we tackle the issue of ambiguity by proposing a system based on deep convolutional neural networks, which is capable of autonomous underground cavity detection beneath urban roads using three-dimensional ground penetrating radar data. First, a basis pursuit-based background filtering algorithm is developed to enhance the visibility of underground objects. The deep convolutional neural network is then established and applied to automatically classify underground objects using the filtered three-dimensional ground penetrating radar data as represented by three types of images: A-, B-, and C-scans. In this study, we utilize a novel two-dimensional grid image consisting of several B- and C-scan images. Cavity, pipe, manhole, and intact features extracted from in situ three-dimensional ground penetrating radar data are used to train the convolutional neural network. The proposed technique is experimentally validated using real three-dimensional ground penetrating radar data obtained from urban roads in Seoul, South Korea.

Download Full-text

Three-dimensional convolutional neural network–based underground object classification using three-dimensional ground penetrating radar data

Structural Health Monitoring ◽

10.1177/1475921720902700 ◽

2020 ◽

Vol 19 (6) ◽

pp. 1884-1893

Author(s):

Shekhroz Khudoyarov ◽

Namgyu Kim ◽

Jong-Jae Lee

Keyword(s):

Neural Network ◽

Deep Learning ◽

Convolutional Neural Network ◽

Ground Penetrating Radar ◽

Three Dimensional ◽

Object Classification ◽

Radar Data ◽

Two Dimensional ◽

Underground Cavities ◽

Ground Penetrating

Ground-penetrating radar is a typical sensor system for analyzing underground facilities such as pipelines and rebars. The technique also can be used to detect an underground cavity, which is a potential sign of urban sinkholes. Multichannel ground-penetrating radar devices are widely used to detect underground cavities thanks to the capacity of informative three-dimensional data. Nevertheless, the three-dimensional ground-penetrating radar data interpretation is unclear and complicated when recognizing underground cavities because similar ground-penetrating radar data reflected from different underground objects are often mixed with the cavities. As it is prevalently known that the deep learning algorithm-based techniques are powerful at image classification, deep learning-based techniques for underground object detection techniques using two-dimensional GPR (ground-penetrating radar) radargrams have been researched upon in recent years. However, spatial information of underground objects can be characterized better in three-dimensional ground-penetrating radar voxel data than in two-dimensional ground-penetrating radar images. Therefore, in this study, a novel underground object classification technique is proposed by applying deep three-dimensional convolutional neural network on three-dimensional ground-penetrating radar data. First, a deep convolutional neural network architecture was developed using three-dimensional convolutional networks for recognizing spatial underground objects such as, pipe, cavity, manhole, and subsoil. The framework of applying the three-dimensional convolutional neural network into three-dimensional ground-penetrating radar data was then proposed and experimentally validated using real three-dimensional ground-penetrating radar data. In order to do that, three-dimensional ground-penetrating radar block data were used to train the developed three-dimensional convolutional neural network and to classify unclassified three-dimensional ground-penetrating radar data collected from urban roads in Seoul, South Korea. The validation results revealed that four underground objects (pipe, cavity, manhole, and subsoil) are successfully classified, and the average classification accuracy was 97%. In addition, a false alarm was rarely indicated.

Download Full-text

3D GPR Image-based UcNet for Enhancing Underground Cavity Detectability

Remote Sensing ◽

10.3390/rs11212545 ◽

2019 ◽

Vol 11 (21) ◽

pp. 2545 ◽

Cited By ~ 3

Author(s):

Man-Sung Kang ◽

Namgyu Kim ◽

Seok Been Im ◽

Jong-Jae Lee ◽

Yun-Kyu An

Keyword(s):

Ground Penetrating Radar ◽

Super Resolution ◽

Data Interpretation ◽

Dependent Data ◽

Test Results ◽

Underground Cavity ◽

Cavity Detection ◽

3D Gpr ◽

Ground Penetrating

This paper proposes a 3D ground penetrating radar (GPR) image-based underground cavity detection network (UcNet) for preventing sinkholes in complex urban roads. UcNet is developed based on convolutional neural network (CNN) incorporated with phase analysis of super-resolution (SR) GPR images. CNNs have been popularly used for automated GPR data classification, because expert-dependent data interpretation of massive GPR data obtained from urban roads is typically cumbersome and time consuming. However, the conventional CNNs often provide misclassification results due to similar GPR features automatically extracted from arbitrary underground objects such as cavities, manholes, gravels, subsoil backgrounds and so on. In particular, non-cavity features are often misclassified as real cavities, which degrades the CNNs’ performance and reliability. UcNet improves underground cavity detectability by generating SR GPR images of the cavities extracted from CNN and analyzing their phase information. The proposed UcNet is experimentally validated using in-situ GPR data collected from complex urban roads in Seoul, South Korea. The validation test results reveal that the underground cavity misclassification is remarkably decreased compared to the conventional CNN ones.

Download Full-text

Defect segmentation: Mapping tunnel lining internal defects with ground penetrating radar data using a convolutional neural network

Construction and Building Materials ◽

10.1016/j.conbuildmat.2021.125658 ◽

2022 ◽

Vol 319 ◽

pp. 125658

Author(s):

Senlin Yang ◽

Zhengfang Wang ◽

Jing Wang ◽

Anthony G. Cohn ◽

Jiaqi Zhang ◽

...

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Ground Penetrating Radar ◽

Radar Data ◽

Tunnel Lining ◽

Internal Defects ◽

Ground Penetrating

Download Full-text

Object Detection in Ground-Penetrating Radar Images Using a Deep Convolutional Neural Network and Image Set Preparation by Migration

International Journal of Geophysics ◽

10.1155/2018/9365184 ◽

2018 ◽

Vol 2018 ◽

pp. 1-8 ◽

Cited By ~ 3

Author(s):

Kazuya Ishitsuka ◽

Shinichiro Iso ◽

Kyosuke Onishi ◽

Toshifumi Matsuoka

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Ground Penetrating Radar ◽

Deep Convolutional Neural Network ◽

Training Data ◽

Neural Network Approach ◽

Radar Images ◽

Image Set ◽

Ground Penetrating ◽

Better Than

Ground-penetrating radar allows the acquisition of many images for investigation of the pavement interior and shallow geological structures. Accordingly, an efficient methodology of detecting objects, such as pipes, reinforcing steel bars, and internal voids, in ground-penetrating radar images is an emerging technology. In this paper, we propose using a deep convolutional neural network to detect characteristic hyperbolic signatures from embedded objects. As a first step, we developed a migration-based method to collect many training data and created 53510 categorized images. We then examined the accuracy of the deep convolutional neural network in detecting the signatures. The accuracy of the classification was 0.945 (94.5%)–0.979 (97.9%) when using several thousands of training images and was much better than the accuracy of the conventional neural network approach. Our results demonstrate the effectiveness of the deep convolutional neural network in detecting characteristic events in ground-penetrating radar images.

Download Full-text

Velocity estimation by the common-reflection-surface (CRS) method: Using ground-penetrating radar data

Geophysics ◽

10.1190/1.2106047 ◽

2005 ◽

Vol 70 (6) ◽

pp. B43-B52 ◽

Cited By ~ 19

Author(s):

Hervé Perroud ◽

Martin Tygel

Keyword(s):

Ground Penetrating Radar ◽

Radar Data ◽

Velocity Estimation ◽

Agricultural Activities ◽

Water Conductivity ◽

Common Reflection Surface ◽

The Common ◽

The Time Domain ◽

Ground Penetrating

In this paper, we describe the use of the common-reflection-surface (CRS) method to estimate velocities from ground-penetrating radar (GPR) data. Applied to multicoverage data, the CRS method provides, as one of its outputs, the time-domain rms velocity map, which is then converted to depth by the familiar Dix algorithm. Combination of the obtained depth-converted velocity map with electrical resistivity in-situ measurements enables us to estimate both water content and water conductivity. These quantities are essential to delineate infiltration of contaminants from the surface after industrial or agricultural activities. The method was applied to GPR data and compared with the classical NMO approach. The results show that the CRS method provides a physically more meaningful velocity field, thus improving the potential of GPR as an investigation tool for environmental studies.

Download Full-text