SUBSURFACE VOIDS DETECTION FROM LIMITED GROUND PENETRATING RADAR DATA USING GENERATIVE ADVERSARIAL NETWORK AND YOLOV5

Guanyi Chen; Xu Bai; Gang Wang; Long Wang; Xuerong Luo; Mingjie Ji; Pengfei Feng; Yang Zhang

doi:10.1109/IGARSS47720.2021.9554954

SUBSURFACE VOIDS DETECTION FROM LIMITED GROUND PENETRATING RADAR DATA USING GENERATIVE ADVERSARIAL NETWORK AND YOLOV5

Guanyi Chen
, Xu Bai
, Gang Wang
, Long Wang
, Xuerong Luo
, Mingjie Ji
, Pengfei Feng
, Yang Zhang

School of Electronics and Information Engineering, Harbin Institute of Technology

Research output: Contribution to conference › Paper › peer-review

15 Scopus citations

Abstract

Recently, preventing road collapse caused by subsurface voids became an urgent problem that needs to be solved in the urban road safety area. While the most study in the field of subsurface object detection by deep learning method has only focused on the objects that can be acquired GPR B-scan data easily. This paper aims to realize the subsurface voids detection under the condition that lacks verified GPR B-Scan data. In this paper, a GPR B-Scan image augmentation method by SinGAN is proposed and the YOLOv5 object detection algorithm is applied correspondingly to detect subsurface voids from both real collected and generated GPR B-Scan data. The detection results show that the proposed technique realized subsurface voids detection in limited verified GPR B-scan data samples and become an inspiration for similar tasks that lacking training samples.

Original language	English
Pages	8600-8603
Number of pages	4
DOIs	https://doi.org/10.1109/IGARSS47720.2021.9554954
State	Published - 2021
Externally published	Yes
Event	2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021 - Brussels, Belgium Duration: 12 Jul 2021 → 16 Jul 2021

Conference

Conference	2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021
Country/Territory	Belgium
City	Brussels
Period	12/07/21 → 16/07/21

Keywords

Generative Adversarial Nets (GANs)
Ground Penetrating Radar (GPR)
Yolov5
deep learning
image augumentation
object detection

Access to Document

10.1109/IGARSS47720.2021.9554954

Cite this

Chen, G., Bai, X., Wang, G., Wang, L., Luo, X., Ji, M., Feng, P., & Zhang, Y. (2021). SUBSURFACE VOIDS DETECTION FROM LIMITED GROUND PENETRATING RADAR DATA USING GENERATIVE ADVERSARIAL NETWORK AND YOLOV5. 8600-8603. Paper presented at 2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021, Brussels, Belgium. https://doi.org/10.1109/IGARSS47720.2021.9554954

@conference{b436632bb4e840339b475d0eb35ccef4,

title = "SUBSURFACE VOIDS DETECTION FROM LIMITED GROUND PENETRATING RADAR DATA USING GENERATIVE ADVERSARIAL NETWORK AND YOLOV5",

abstract = "Recently, preventing road collapse caused by subsurface voids became an urgent problem that needs to be solved in the urban road safety area. While the most study in the field of subsurface object detection by deep learning method has only focused on the objects that can be acquired GPR B-scan data easily. This paper aims to realize the subsurface voids detection under the condition that lacks verified GPR B-Scan data. In this paper, a GPR B-Scan image augmentation method by SinGAN is proposed and the YOLOv5 object detection algorithm is applied correspondingly to detect subsurface voids from both real collected and generated GPR B-Scan data. The detection results show that the proposed technique realized subsurface voids detection in limited verified GPR B-scan data samples and become an inspiration for similar tasks that lacking training samples.",

keywords = "Generative Adversarial Nets (GANs), Ground Penetrating Radar (GPR), Yolov5, deep learning, image augumentation, object detection",

author = "Guanyi Chen and Xu Bai and Gang Wang and Long Wang and Xuerong Luo and Mingjie Ji and Pengfei Feng and Yang Zhang",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021 ; Conference date: 12-07-2021 Through 16-07-2021",

year = "2021",

doi = "10.1109/IGARSS47720.2021.9554954",

language = "英语",

pages = "8600--8603",

}

Chen, G, Bai, X, Wang, G, Wang, L, Luo, X, Ji, M, Feng, P & Zhang, Y 2021, 'SUBSURFACE VOIDS DETECTION FROM LIMITED GROUND PENETRATING RADAR DATA USING GENERATIVE ADVERSARIAL NETWORK AND YOLOV5', Paper presented at 2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021, Brussels, Belgium, 12/07/21 - 16/07/21 pp. 8600-8603. https://doi.org/10.1109/IGARSS47720.2021.9554954

TY - CONF

T1 - SUBSURFACE VOIDS DETECTION FROM LIMITED GROUND PENETRATING RADAR DATA USING GENERATIVE ADVERSARIAL NETWORK AND YOLOV5

AU - Chen, Guanyi

AU - Bai, Xu

AU - Wang, Gang

AU - Wang, Long

AU - Luo, Xuerong

AU - Ji, Mingjie

AU - Feng, Pengfei

AU - Zhang, Yang

PY - 2021

Y1 - 2021

N2 - Recently, preventing road collapse caused by subsurface voids became an urgent problem that needs to be solved in the urban road safety area. While the most study in the field of subsurface object detection by deep learning method has only focused on the objects that can be acquired GPR B-scan data easily. This paper aims to realize the subsurface voids detection under the condition that lacks verified GPR B-Scan data. In this paper, a GPR B-Scan image augmentation method by SinGAN is proposed and the YOLOv5 object detection algorithm is applied correspondingly to detect subsurface voids from both real collected and generated GPR B-Scan data. The detection results show that the proposed technique realized subsurface voids detection in limited verified GPR B-scan data samples and become an inspiration for similar tasks that lacking training samples.

AB - Recently, preventing road collapse caused by subsurface voids became an urgent problem that needs to be solved in the urban road safety area. While the most study in the field of subsurface object detection by deep learning method has only focused on the objects that can be acquired GPR B-scan data easily. This paper aims to realize the subsurface voids detection under the condition that lacks verified GPR B-Scan data. In this paper, a GPR B-Scan image augmentation method by SinGAN is proposed and the YOLOv5 object detection algorithm is applied correspondingly to detect subsurface voids from both real collected and generated GPR B-Scan data. The detection results show that the proposed technique realized subsurface voids detection in limited verified GPR B-scan data samples and become an inspiration for similar tasks that lacking training samples.

KW - Generative Adversarial Nets (GANs)

KW - Ground Penetrating Radar (GPR)

KW - Yolov5

KW - deep learning

KW - image augumentation

KW - object detection

UR - https://www.scopus.com/pages/publications/85129841482

U2 - 10.1109/IGARSS47720.2021.9554954

DO - 10.1109/IGARSS47720.2021.9554954

M3 - 论文

AN - SCOPUS:85129841482

SP - 8600

EP - 8603

T2 - 2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021

Y2 - 12 July 2021 through 16 July 2021

ER -

SUBSURFACE VOIDS DETECTION FROM LIMITED GROUND PENETRATING RADAR DATA USING GENERATIVE ADVERSARIAL NETWORK AND YOLOV5

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this