Recognition for underground voids in C-scans based on GMM-HMM

doi:10.23919/JSEE.2024.000093

Journal of Systems Engineering and Electronics ›› 2025, Vol. 36 ›› Issue (1): 82-94.doi: 10.23919/JSEE.2024.000093

• DEFENCE ELECTRONICS TECHNOLOGY • Previous Articles

Recognition for underground voids in C-scans based on GMM-HMM

Xu BAI(), Yuhao LI(), Shizeng GUO(), Jinlong LIU(), Zhitao WEN(), Hongrui LI(), Jiayan ZHANG()

School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin 150001, China

Received:2023-05-09 Accepted:2024-07-20 Online:2025-02-18 Published:2025-03-18
Contact: Shizeng GUO E-mail:x_bai@hit.edu.cn;hitliyh@foxmail.com;21S105221@stu.hit.edu.cn;1312471902@qq.com;wzt_1998@163.com;1336144982@qq.com;jyzhang@hit.edu.cn
About author:
BAI Xu was born in 1974. He received his B.S., M.S., and Ph.D. degrees from the School of Electronics and Infornation Engineering, Harbin Institute of Technology, Harbin, China, in 1997, 2004 and 2008. He is currently an Associate professor with the Communication Research Centre of Harbin Institute of Technology. His research interests include image processing, wireless communication, signal processing in ground penetrating radar, and wireless sensor network. E-mail: x_bai@hit.edu.cn

LI Yuhao was born in 1998. He received his B.S. degree from the School of Electronics and Infornation Engineering, Harbin Institute of Technology, Harbin, China, in 2021. He is pursuing his M.S. degree in electronics and infornation engineering with Harbin Institute of Technology. His research interests include system modeling and simulation and machine learning. E-mail: hitliyh@foxmail.com

GUO Shizeng was born in 1965. He received his M.S. degree from the School of Electronics and Infornation Engineering, Harbin Institute of Technology, Harbin, China, in 2005. Currently he is an associate professor with Harbin Institute of Technology. His research interests include pass-switching technology and dedicated data chain testing technology. E-mail: 21S105221@stu.hit.edu.cn

LIU Jinlong was born in 1998. He received his M.S. degree from the School of Electronics and Infornation Engineering, Harbin Institute of Technology, Harbin, China, in 2023. His research interests are hardware circuit design and deep learning. E-mail: 1312471902@qq.com

WEN Zhitao was born in 1998. He received his M.S. degree from the School of Electronics and Infornation Engineering, Harbin Institute of Technology, Harbin, China, in 2023. His research interests are wireless communication and ground penetrating radar signal processing. E-mail: wzt_1998@163.com

LI Hongrui was born in 1998. He received his B.S. degree from the School of Electronics and Infornation Engineering, Harbin Institute of Technology, Harbin, China, in 2021. He is pursuing his M.S. degree in Electronics and Infornation Engineering, Harbin Institute of Technology. His research interests are ground penetrating radar signal processing and deep learning. E-mail: 1336144982@qq.com

ZHANG Jiayan was born in 1974. He received his Ph.D. degree from the School of Electronics and Infornation Engineering, Harbin Institute of Technology, Harbin, China, in 2007. His research interests are artificial intelligence, high speed signal processing and ground penetrating radar.E-mail: jyzhang@hit.edu.cn
Supported by:
This work was supported by the National Natural Science Foundation of China (62071147).

Abstract

Abstract:

Ground penetrating radar (GPR), as a fast, efficient, and non-destructive detection device, holds great potential for the detection of shallow subsurface environments, such as urban road subsurface monitoring. However, the interpretation of GPR echo images often relies on manual recognition by experienced engineers. In order to address the automatic interpretation of cavity targets in GPR echo images, a recognition-algorithm based on Gaussian mixed model-hidden Markov model (GMM-HMM) is proposed, which can recognize three dimensional (3D) underground voids automatically. First, energy detection on the echo images is performed, whereby the data is pre-processed and pre-filtered. Then, edge histogram descriptor (EHD), histogram of oriented gradient (HOG), and Log-Gabor filters are used to extract features from the images. The traditional method can only be applied to 2D images and pre-processing is required for C-scan images. Finally, the aggregated features are fed into the GMM-HMM for classification and compared with two other methods, long short-term memory (LSTM) and gate recurrent unit (GRU). By testing on a simulated dataset, an accuracy rate of 90% is obtained, demonstrating the effectiveness and efficiency of our proposed method.

Key words: ground penetrating rader (GPR), recognition, edge histogram descriptor (EHD), histogram of oriented gradient (HOG), Log-Gabor filter

Xu BAI, Yuhao LI, Shizeng GUO, Jinlong LIU, Zhitao WEN, Hongrui LI, Jiayan ZHANG. Recognition for underground voids in C-scans based on GMM-HMM[J]. Journal of Systems Engineering and Electronics, 2025, 36(1): 82-94.

Figures/Tables 23

Fig 1

Fig 2

Fig 3

Fig 4

Fig 5

Fig 6

Fig 7

Fig 8

Fig 9

Fig 10

Fig 11

Fig 12

Fig 13

Fig 14

Fig 15

Fig 16

Fig 17

Fig 18

Table 1

Table 2

Table 3

Table 4

Table 5

References 31

1	XU Z S, JIAN S K, TAN T. Application of ground penetrating radar (GPR) in detection of road void cavity disease. Chinese Journal of Engineering Geophysics, 2019(1): 110−128. (in Chinese)
2	WU Y Y, XIE Y Z, CAO H, et al Test methods for performance of protective devices excited by conducted transient electromagnetic disturbance. IEEE Letters on Electromagnetic Compatibility Practice and Applications, 2023, 5 (3): 72- 76. doi: 10.1109/LEMCPA.2023.3290384
3	TRAVASSOS X L, AVILA S L, IDA N Artificial neural networks and machine learning techniques applied to ground penetrating radar: a review. Applied Computing and Informatics, 2020, 17 (2): 296- 308.
4	FEDOROV M P, FEDOROVA L L, OMELYANAENKO A V. Investigation of the Lena river ice cover by GPR from helicopter. Proc. of the IEEE 14th International Conference on Ground Penetrating Radar, 2012: 733−736.
5	LIU H, DING F, LI J H, et al Improved detection of buried elongated targets by dual-polarization GPR. IEEE Geoscience and Remote Sensing Letters, 2023, 20, 3501705.
6	SHI X H, CHENG D D, SONG Z Y, et al. A real-time method for landmine detection using vehicle array GPR. Proc. of the IEEE 17th International Conference on Ground Penetrating Radar, 2018. DOI: 10.1109/ICGPR.2018.8441584.
7	KUMAWAT P, KHATRI A, NAGARIA B. Comparative analysis of offline handwriting recognition using invariant moments with HMM and combined SVM-HMM classifier. Proc. of the IEEE International Conference on Communication Systems and Network Technologies, 2013: 140−143.
8	CASTELLINI A, MASILLO F, AZZALINI D, et al Adversarial data augmentation for HMM-based anomaly detection. IEEE Trans. on Pattern Analysis and Machine Intelligence, 2023, 45 (12): 14131- 14143. doi: 10.1109/TPAMI.2023.3303099
9	JIANG Z, HUANG A, QI G Q, et al A framework of travel mode identification fusing deep learning and map-matching algorithm. IEEE Trans. on Intelligent Transportation Systems, 2023, 24 (6): 6401- 6415. doi: 10.1109/TITS.2023.3250660
10	TING W. An acoustic recognition model for english speech based on improved HMM algorithm. Proc. of the IEEE 11th International Conference on Measuring Technology and Mechatronics Automation, 2019: 729−732.
11	LEE S, SHIN Y, KIM M, et al IR-UWB radar-based contactless silent speech recognition of vowels, consonants, words, and phrases. IEEE Access, 2023, 11, 144844- 144859. doi: 10.1109/ACCESS.2023.3344177
12	TIAN J C, YU J W, WENG C, et al Integrating lattice-free MMI into end-to-end speech recognition. IEEE/ACM Trans. on Audio, Speech, and Language Processing, 2022, 31, 25- 38.
13	DAVID R, SÖFFKER D A study on a HMM-based state machine approach for lane changing behavior recognition. IEEE Access, 2022, 10, 122954- 122964. doi: 10.1109/ACCESS.2022.3224012
14	ABUJARAD F, NADIM G, OMAR A. Wavelet packets for GPR detection of non-metallic anti-personnel land mines based on higher-order-statistic. Proc. of the 3rd International Workshop on Advance Ground Penetrating Radar, 2005: 21−24.
15	FRIGUI H, MISSAOUI O, GADER P Landmine detection using discrete hidden Markov models with Gabor features. Proc. of the 12th Detection and Remediation Technologies for Mines and Minelike Targets, 2007, 674- 683.
16	MISSAOUI O, FRIGUI H, GADER P Land-mine detection with ground-penetrating radar using multistream discrete hidden Markov models. IEEE Trans. on Geoscience and Remote Sensing, 2010, 49 (6): 2080- 2099.
17	ZHANG X P, YUKSEL S E, GADER P, et al. Simultaneous feature and HMM model learning for landmine detection using ground penetrating radar. Proc. of the IEEE IAPR Workshop on Pattern Recognition in Remote Sensing, 2010. DOI: 10.1109/PRRS.2010.5742805.
18	RATTO C R, MORTON K D, COLLINS L M, et al. A hidden Markov context model for GPR-based landmine detection incorporating stick-breaking priors. Proc. of the IEEE International Geoscience and Remote Sensing Symposium, 2011: 874−877.
19	WILLIAMS R M, RAY L E, LEVER J H, et al Crevasse detection in ice sheets using ground penetrating radar and machine learning. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7 (12): 4836- 4848. doi: 10.1109/JSTARS.2014.2332872
20	MANANDHAR A, TORRIONE P A, COLLINS L M, et al Multiple-instance hidden Markov model for GPR-based landmine detection. IEEE Trans. on Geoscience and Remote Sensing, 2014, 53 (4): 1737- 1745.
21	YUKSEL S E, BOLTON J, GADER P Multiple-instance hidden Markov models with applications to landmine detection. IEEE Trans. on Geoscience and Remote Sensing, 2015, 53 (12): 6766- 6775. doi: 10.1109/TGRS.2015.2447576
22	DALAL N, TRIGGS B Histograms of oriented gradients for human detection. Proc. of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2005, 1, 886- 893.
23	LEE K L, MOKJI M M. Automatic target detection in GPR images using histogram of oriented gradients (HOG). Proc. of the IEEE 2nd International Conference on Electronic design, 2014: 181−186.
24	KAREM A, FRIGUI H. A multiple instance learning approach for landmine detection using ground penetrating radar. Proc. of the IEEE International Geoscience and Remote Sensing Symposium, 2011: 878−881.
25	KOBAYASHI M, NAKANO K. Dirichlet process crescent-signal mixture model for ground-penetrating radar signals. Proc. of the IEEE 40th Annual Conference on Industrial Electronics Society, 2014: 3431−3437.
26	ROHMAN B P A, NISHIMOTO M. Near-surface soil water content estimation using UWB-GPR based on selective sparse representation. Proc. of the IEEE Sensors Applications Symposium, 2018. DOI: 10.1109/SAS.2018.8336717.
27	SAVITA S J, PALLAVI A. Modeling of GPR using gprMax simulation. Proc. of the IEEE International Conference on Distributed Computing and Electrical Circuits and Electronics, 2022. DOI: 10.1109/ICDCECE53908.2022.9792883.
28	LI N S, WU R B, LI H F, et al MV-GPRNet: multi-view subsurface defect detection network for airport runway inspection based on GPR. Remote Sensing, 2022, 14 (18): 4472. doi: 10.3390/rs14184472
29	LIU L, YU H, XU H, et al Underground object classification using deep 3-D convolutional networks and multiple mirror encoding for GPR data. IEEE Geoscience and Remote Sensing Letters, 2022, 19, 4021705.
30	ZHOU Y M, LAI W W L. Leakage detection using ground penetrating radar C-scan based on 3D fuzzy C-means clustering. Proc. of the IEEE 12th International Workshop on Advanced Ground Penetrating Radar, 2023. DOI: 10.1109/IWAGPR57138.2023.10329148.
31	DAI Q Q, LEE Y H, SUN H H, et al 3DInvNet: a deep learning-based 3D ground-penetrating radar data inversion. IEEE Trans. on Geoscience and Remote Sensing, 2023, 61, 5105016.

Parameter	Value
Model size/m	(1.5, 1.5, 1.5)
Step length/m	(0.01, 0.01, 0.01)
Surface relative dielectric constant	4
Surface conductivity/(S/m)	0.005
Surface relative permeability	1
Surface magnetic loss/(Ω/m)	0
Base relative dielectric constant	5
Base conductivity/(S/m)	0.05
Base relative permeability	1
Base magnetic loss/(Ω/m)	0
Waveform	Riker
Maximum amplitude/A	1
Radius of spherical voids/m	0.1−0.7
Size of cubic voids/m	(0.1−0.7, 0.1−0.7, 0.1−0.7)
B-scans in single C-scan	14

Parameter	Value
Sand proportion	0.5
Clay proportion	0.5
Clod density/(g/cm³)	2.0
Grit density	2.66 0.001−0.25
Water content	2.66 0.001−0.25
Hybrid material amount Impurity type	5−50 PEC and others

Feature extraction	Method	Accuracy
Feature extraction	Method	Block interference/%	Gaussian noise/%
EHD	2D image sequence	96.3	53.7
EHD	Mean	92.6	70.4
HOG	2D image sequence	63.0	68.5
HOG	Mean	92.6	53.7
Log-Gabor	2D image sequence	59.3	75.9
Log-Gabor	Mean	53.7	83.3

Feature extraction	Method	Accuracy
Feature extraction	Method	Block interference/%	Gaussian noise/%
EHD	2D image sequence	98.2	57.7
EHD	Mean	94.1	76.3
HOG	2D image sequence	90.2	50.8
HOG	Mean	94.3	58.7
Log-Gabor	2D image sequence	83.3	98.9
Log-Gabor	Mean	96.6	98.4

Feature extraction	Method	Accuracy
Feature extraction	Method	Block interference/%	Gaussian noise/%
EHD	2D image sequence	98.1	56.4
EHD	Mean	95.0	76.6
HOG	2D image sequence	90.7	50.0
HOG	Mean	93.1	59.4
Log-Gabor	2D image sequence	89.8	99.1
Log-Gabor	Mean	96.0	97.4

Recognition for underground voids in C-scans based on GMM-HMM

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

Share this article

Figures/Tables 23

References 31

Related Articles 15

Recommended Articles

Metrics

Comments

[1]	Hongcheng YIN, Hua YAN. Parametric modeling and applications of target scattering centers: a review [J]. Journal of Systems Engineering and Electronics, 2024, 35(6): 1411-1427.
[2]	Xiaobo DUAN, Qiucen FAN, Wenhao BI, An ZHANG. Belief exponential divergence for D-S evidence theory and its application in multi-source information fusion [J]. Journal of Systems Engineering and Electronics, 2024, 35(6): 1454-1468.
[3]	Ruihui PENG, Xingrui WU, Guohong WANG, Dianxing SUN, Zhong YANG, Hongwen LI. Intelligent recognition and information extraction of radar complex jamming based on time-frequency features [J]. Journal of Systems Engineering and Electronics, 2024, 35(5): 1148-1166.
[4]	Haibin WANG, Xin GUAN, Xiao YI, Guidong SUN. Heterogeneous information fusion recognition method based on belief rule structure [J]. Journal of Systems Engineering and Electronics, 2024, 35(4): 955-964.
[5]	Xueling YANG, Gong ZHANG, Hu SONG. Ship recognition based on HRRP via multi-scale sparse preserving method [J]. Journal of Systems Engineering and Electronics, 2024, 35(3): 599-608.
[6]	Hao DU, Wei WANG, Xuerao WANG, Jingqiu ZUO, Yuanda WANG. Scene image recognition with knowledge transfer for drone navigation [J]. Journal of Systems Engineering and Electronics, 2023, 34(5): 1309-1318.
[7]	Yasong LUO, Jianghu XU, Chengxu FENG, Kun ZHANG. An accurate detection algorithm for time backtracked projectile-induced water columns based on the improved YOLO network [J]. Journal of Systems Engineering and Electronics, 2023, 34(4): 981-991.
[8]	Yunxiu ZENG, Kai XU. Recognition and interfere deceptive behavior based on inverse reinforcement learning and game theory [J]. Journal of Systems Engineering and Electronics, 2023, 34(2): 270-288.
[9]	Hao DU, Wei WANG, Xuerao WANG, Yuanda WANG. Autonomous landing scene recognition based on transfer learning for drones [J]. Journal of Systems Engineering and Electronics, 2023, 34(1): 28-35.
[10]	Zhenyu HE, Xiaodong ZHUGE, Junxiang WANG, Shihao YU, Yongjun XIE, Yuxiong ZHAO. Half space object classification via incident angle based fusion of radar and infrared sensors [J]. Journal of Systems Engineering and Electronics, 2022, 33(5): 1025-1031.
[11]	Lingchi GE, Min FANG, Haikun LI, Bo CHEN. Label correlation for partial label learning [J]. Journal of Systems Engineering and Electronics, 2022, 33(5): 1043-1051.
[12]	Dong FU, Xiangjun LI, Weihua MOU, Ming MA, Gang OU. Navigation jamming signal recognition based on long short-term memory neural networks [J]. Journal of Systems Engineering and Electronics, 2022, 33(4): 835-844.
[13]	Ying YUAN, Feng YU, Yang CHEN, Niancheng ZHANG. A method to realize NAVSOP by utilizing GNSS authorized signals [J]. Journal of Systems Engineering and Electronics, 2021, 32(5): 1232-1245.
[14]	Xia WU, Yan LI, Yongjian SUN, Alei CHEN, Jianwen CHEN, Jianchao MA, Hao CHEN. Investigation of MAS structure and intelligent⁺ information processing mechanism of hypersonic target detection and recognition system [J]. Journal of Systems Engineering and Electronics, 2020, 31(6): 1105-1115.
[15]	Pengcheng GUO, Zheng LIU, Jingjing WANG. Radar group target recognition based on HRRPs and weighted mean shift clustering [J]. Journal of Systems Engineering and Electronics, 2020, 31(6): 1152-1159.