Automatic detection of targets under airport pavement based on channel and spatial attention

doi:10.11772/j.issn.1001-9081.2022020168

Journal of Computer Applications ›› 2023, Vol. 43 ›› Issue (3): 930-935.DOI: 10.11772/j.issn.1001-9081.2022020168

• Multimedia computing and computer simulation • Previous Articles

Automatic detection of targets under airport pavement based on channel and spatial attention

Haifeng LI¹, Fan ZHANG¹, Minnan PIAO¹(), Huaichao WANG¹, Nansha LI¹, Zhongcheng GUI²

^1.College of Computer Science and Technology，Civil Aviation University of China，Tianjin 300300，China
^2.Chengdu Guimu Robot Company Limited，Chengdu Sichuan 610101，China

Received:2022-02-16 Revised:2022-05-12 Accepted:2022-05-13 Online:2022-08-16 Published:2023-03-10
Contact: Minnan PIAO
About author:LI Haifeng， born in 1984， Ph. D.， associate professor. His research interests include robot environmental perception， computer vision.
ZHANG Fan， born in 1998， M. S. candidate. His research interests include computer vision， image processing.
WANG Huaichao， born in 1984， Ph. D.， lecturer. His research interests include computer vision.
LI Nansha， born in 1996， Ph. D. candidate. Her research interests include computer vision.
GUI Zhongcheng， born in 1979， Ph.D.， professor of engineering. His research interests include intelligent robot.
Supported by:
National Key Research and Development Program of China(2019YFB1310400);Scientific Research Program of Tianjin Municipal Education Commission(2021KJ036)

基于通道和空间注意力的机场道面地下目标自动检测

李海丰¹, 张凡¹, 朴敏楠¹(), 王怀超¹, 李南莎¹, 桂仲成²

^1.中国民航大学计算机科学与技术学院，天津 300300
^2.成都圭目机器人有限公司，成都 610101

通讯作者: 朴敏楠
作者简介:李海丰（1984—），男，内蒙古通辽人，副教授，博士，CCF会员，主要研究方向：机器人环境感知、计算机视觉
张凡（1998—），男，四川巴中人，硕士研究生，主要研究方向：计算机视觉、图像处理
朴敏楠（1993—），女，吉林四平人，讲师，博士，主要研究方向：机器人智能感知与决策
王怀超（1984—），男，天津人，讲师，博士，主要研究方向：计算机视觉
李南莎（1996—），女，湖北咸宁人，博士研究生，主要研究方向：计算机视觉
桂仲成（1979—），男，安徽六安人，教授级高级工程师，博士，主要研究方向：智能机器人。
基金资助:
国家重点研发计划项目(2019YFB1310400);天津市教委科研计划项目(2021KJ036)

Abstract

Abstract:

In the task of detecting targets under airport pavement， B-scan maps generated by Ground Penetrating Radar （GPR） have complex backgrounds and lots of noise， especially a single B-scan map cannot reflect the complete information of an underground target. To solve these problems， a Three-Dimensional Channel and Spatial Attention UNet （3D-CSA-UNet） model was established to automatically detect the underground targets. Firstly， a Three-Dimensional Channel and Spatial parallel attention Block （3D-CS-Block） was designed to make the model focus on the underground target information in radar C-scan and suppress the interference of backgrounds and noise. Secondly， in order to enhance the capability of 3D-CS-Block in feature extraction， a multi-scale 3D segmentation model was designed to extract feature maps of different sizes from the radar C-scan. Finally， the cross-entropy loss function was employed to calculate the loss value of feature map under each scale to improve the detection accuracy of the model. On a real dataset of targets under airport pavement， compared with 3D-Fully Convolutional Network （3D-FCN）， 3D-UNet and other algorithms， 3D-CSA-UNet has the average F1 score in terms of the pixel level segmentation for void， rebar and parallel rebar targets increased by at last 12.33， 9.05 and 11.05 percentage points. Experimental results show that 3D-CSA-UNet can meet the real engineering requirements well.

Key words: Ground Penetrating Radar (GPR), target detection, Convolutional Neural Network (CNN), channel attention, spatial attention, feature extraction

摘要：

针对机场道面地下目标检测任务中，探地雷达（GPR）生成的B-scan图背景复杂、包含大量噪声，尤其是单个B-scan图不能反映地下目标的完整信息等问题，构建一种三维通道和空间注意力的UNet（3D-CSA-UNet）模型对地下目标进行自动检测。首先，设计三维通道和空间注意力并行模块（3D-CS-Block），使模型重点关注雷达C-scan中的地下目标信息，抑制背景和噪声的干扰；其次，设计多尺度的三维分割模型从雷达C-scan中提取不同大小的特征图，以增强3D-CS-Block提取目标特征的能力；最后，使用交叉熵损失函数计算每个尺度下特征图的损失值，从而提高模型的检测精度。在采集的实际机场道面地下目标数据集上，相较于3D-FCN、3D-UNet等模型，3D-CSA-UNet对于脱空、钢筋和钢筋平行目标预测的平均F1至少提高12.33、9.05、11.05个百分点。实验结果表明，3D-CSA-UNet可以较好地满足工程实际要求。

关键词: 探地雷达, 目标检测, 卷积神经网络, 通道注意力, 空间注意力, 特征提取

CLC Number:

TP391.41

Haifeng LI, Fan ZHANG, Minnan PIAO, Huaichao WANG, Nansha LI, Zhongcheng GUI. Automatic detection of targets under airport pavement based on channel and spatial attention[J]. Journal of Computer Applications, 2023, 43(3): 930-935.

李海丰, 张凡, 朴敏楠, 王怀超, 李南莎, 桂仲成. 基于通道和空间注意力的机场道面地下目标自动检测[J]. 《计算机应用》唯一官方网站, 2023, 43(3): 930-935.

Figures/Tables 10

References 21

1	杨必胜，宗泽亮，陈驰，等. 车载探地雷达地下目标实时探测法［J］. 测绘学报， 2020， 49（7）： 874-882.
	YANG B S， ZONG Z L， CHEN C， et al. Real time approach for underground objects detection from vehicle-borne ground penetrating radar［J］. Acta Geodaetica et Cartographica Sinica， 2020， 49（7）： 874-882.
2	黄晓惠. 透水混凝土整体路面施工质量控制与检测技术研究［D］. 绵阳：西南科技大学， 2021： 47-52.
	HUANG X H. Research on construction quality control and detection technology for pervious concrete monolithic pavement［D］. Mianyang： Southwest University of Science and Technology， 2021： 47-52.
3	郭云飞. 机场水泥混凝土道面脱空注浆修复评价研究［D］. 郑州：郑州大学， 2021： 46-56.
	GUO Y F. Research on evaluation of void grouting repair for airport cement concrete pavement［D］. Zhengzhou： Zhengzhou University， 2021： 46-56.
4	LIU K D， GAO Z B， CHEN F C， et al. Application of airport pavement structure safety detection and intelligent recognition technology［C］// Proceedings of the 4th International Symposium on Traffic Transportation and Civil Architecture. Piscataway： IEEE， 2021： 143-146. 10.1109/isttca53489.2021.9654781
5	何炜琨. 基于探地雷达的机场场道质量监测关键技术研究［D］. 天津：天津大学， 2012： 12-17.
	HE W K. Enabling techniques for runways quality surveillence via ground penetrating radar［D］. Tianjin： Tianjin University， 2012： 12-17.
6	曹芸茜，何炜琨. 场道钢筋加固区的探地雷达信息处理算法及系统研究［J］. 信号处理， 2019， 35（3）：351-360.
	CAO Y Q， HE W K. Research of ground penetrating radar information processing algorithm and system for rebar-reinforced zone in runways［J］. Journal of Signal Processing， 2019， 35（3）： 351-360.
7	何炜琨，吴仁彪，刘家学. 基于探地雷达的机场场道脱空层检测及厚度估计［J］. 信号处理， 2011， 27（10）：1547-1551. 10.3969/j.issn.1003-0530.2011.10.016
	HE W K， WU R B， LIU J X. Void-layer detection and depth determination in runways based on GPR［J］. Signal Processing， 2011， 27（10）： 1547-1551. 10.3969/j.issn.1003-0530.2011.10.016
8	ZOU L L， YI L， SATO M. On the use of lateral wave for the interlayer debonding detecting in an asphalt airport pavement using a multistatic GPR system［J］. IEEE Transactions on Geoscience and Remote Sensing， 2020， 58（6）： 4215-4224. 10.1109/TGRS.2019.2961772
9	PHAM M T， LEFÈVRE S. Buried object detection from B-scan ground penetrating radar data using Faster-RCNN［C］// Proceedings of the 2018 IEEE International Geoscience and Remote Sensing Symposium. Piscataway： IEEE， 2018： 6804-6807. 10.1109/igarss.2018.8517683
10	李海丰，赵碧帆，侯谨毅，等. 基于自适应双阈值的地下目标自动检测算法［J］. 计算机应用， 2022， 42（4）：1275-1283.
	LI H F， ZHAO B F， HOU J Y， et al. Automatic detection algorithm for underground target based on adaptive double thresholds［J］. Journal of Computer Applications， 2022， 42（4）： 1275-1283.
11	DINH K， GUCUNSKI N， DUONG T H. An algorithm for automatic localization and detection of rebars from GPR data of concrete bridge decks［J］. Automation in Construction， 2018， 89： 292-298. 10.1016/j.autcon.2018.02.017
12	CAI B， CHENG E K， LIANG P P， et al. Ghost-Light-3DNet： efficient network for heart segmentation［C］// Proceedings of the IEEE 18th International Symposium on Biomedical Imaging. Piscataway： IEEE， 2021： 1704-1708. 10.1109/isbi48211.2021.9433974
13	XU X N， ZHOU F G， LIU B， et al. Efficient multiple organ localization in CT image using 3D region proposal network［J］. IEEE Transactions on Medical Imaging， 2019， 38（8）： 1885-1898. 10.1109/tmi.2019.2894854
14	AL-HAMMADI M， MUHAMMAD G， ABDUL W， et al. Hand gesture recognition for sign language using 3DCNN［J］. IEEE Access， 2020， 8： 79491-79509. 10.1109/access.2020.2990434
15	LI H F， LI N S， WU R B， et al. GPR-RCNN： an algorithm of subsurface defect detection for airport runway based on GPR［J］. IEEE Robotics and Automation Letters， 2021， 6（2）： 3001-3008. 10.1109/lra.2021.3062599
16	HU J， SHEN L， ALBANIE S， et al. Squeeze-and-excitation networks［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2020， 42（8）： 2011-2023. 10.1109/tpami.2019.2913372
17	RONNEBERGER O， FISCHER P， BROX T. U-net： convolutional networks for biomedical image segmentation［C］// Proceedings of the 2015 International Conference on Medical Image Computing and Computer-Assisted Intervention， LNCS 9351. Cham： Springer， 2015： 234-241.
18	LIU W， ANGUELOY D， ERHAN D， et al. SSD： single shot multiBox detector［C］// Proceedings of the 2016 European Conference on Computer Vision， LNCS 9905. Cham： Springer， 2016： 21-37.
19	DONG S D. A separate 3D-SegNet based on priority queue for brain tumor segmentation［C］// Proceedings of the 12th International Conference on Intelligent Human-Machine Systems and Cybernetics. Piscataway： IEEE， 2020： 140-143. 10.1109/ihmsc49165.2020.10110
20	SHEN Y， FANG Z J， GAO Y B， et al. Coronary arteries segmentation based on 3D FCN with attention gate and level set function［J］. IEEE Access， 2019， 7： 42826-42835. 10.1109/access.2019.2908039
21	CHANG J， ZHANG X C， CHANG J， et al. Brain tumor segmentation based on 3D UNet with multi-class focal loss［C］// Proceedings of the 11th International Congress on Image and Signal Processing， BioMedical Engineering and Informatics. Piscataway： IEEE， 2018： 1-5. 10.1109/cisp-bmei.2018.8633056

数据类型	训练样本数	验证样本数	测试样本数
B-scan	1 722	316	319
C-scan	150	30	35

数据类型	训练样本数	验证样本数	测试样本数
B-scan	1 722	316	319
C-scan	150	30	35

模型	脱空			钢筋			钢筋平行
模型	P	R	F1	P	R	F1	P	R	F1
3D-FCN	56.54	53.72	55.09	68.08	71.43	69.71	74.81	63.35	68.61
3D-SegNet	59.61	52.45	55.93	67.30	73.72	70.36	60.80	79.80	69.01
3D-UNet	66.02	66.18	66.09	81.20	67.12	73.50	65.42	87.23	74.77
3D-CSA-UNet	68.56	66.33	67.42	78.71	78.81	78.76	70.36	91.78	79.66

模型	脱空			钢筋			钢筋平行
模型	P	R	F1	P	R	F1	P	R	F1
3D-FCN	56.54	53.72	55.09	68.08	71.43	69.71	74.81	63.35	68.61
3D-SegNet	59.61	52.45	55.93	67.30	73.72	70.36	60.80	79.80	69.01
3D-UNet	66.02	66.18	66.09	81.20	67.12	73.50	65.42	87.23	74.77
3D-CSA-UNet	68.56	66.33	67.42	78.71	78.81	78.76	70.36	91.78	79.66

实验编号	脱空			钢筋			钢筋平行
实验编号	P	R	F1	P	R	F1	P	R	F1
A	61.16	63.29	62.21	75.42	74.59	75.00	64.07	83.95	72.67
B	63.44	65.53	64.47	71.48	78.43	74.79	65.61	90.79	76.17
C	69.33	64.17	66.65	77.49	74.99	76.22	69.76	84.30	76.34
D	72.39	64.42	68.17	78.10	78.42	78.26	67.83	85.71	75.73
E	76.20	55.86	64.47	74.86	76.04	75.45	61.43	92.32	73.77
F	68.56	66.33	67.42	78.71	78.81	78.76	70.36	91.78	79.66

Automatic detection of targets under airport pavement based on channel and spatial attention

基于通道和空间注意力的机场道面地下目标自动检测

RichHTML

PDF

PDF (Mobile)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 21

Related Articles 15

Recommended Articles

Metrics

[1]	Yongxiang GU, Xin LAN, Boyi FU, Xiaolin QIN. Object detection algorithm for remote sensing images based on geometric adaptation and global perception [J]. Journal of Computer Applications, 2023, 43(3): 916-922.
[2]	Tao PENG, Yalong KANG, Feng YU, Zili ZHANG, Junping LIU, Xinrong HU, Ruhan HE, Li LI. Pedestrian trajectory prediction based on multi-head soft attention graph convolutional network [J]. Journal of Computer Applications, 2023, 43(3): 736-743.
[3]	Ping WANG, Nan CHEN, Lei LU. Fall detection algorithm based on scene prior and attention guidance [J]. Journal of Computer Applications, 2023, 43(2): 529-535.
[4]	Qing JIA, Laihua WANG, Weisheng WANG. Anomaly detection in video via independently recurrent neural network and variational autoencoder network [J]. Journal of Computer Applications, 2023, 43(2): 507-513.
[5]	Li’an ZHU, Hong ZHANG. Nonhomogeneous image dehazing based on dual-branch conditional generative adversarial network [J]. Journal of Computer Applications, 2023, 43(2): 567-574.
[6]	Ranyan NI, Yi ZHANG. Action recognition method based on video spatio-temporal features [J]. Journal of Computer Applications, 2023, 43(2): 521-528.
[7]	Zhijun SHEN, Lina MU, Jing GAO, Yuanhang SHI, Zhiqiang LIU. Review of fine-grained image categorization [J]. Journal of Computer Applications, 2023, 43(1): 51-60.
[8]	Haiyan CHEN, Mingming JIA, Wenli ZHAO, Chanfei WANG. Target detection of Ochotona curzoniae based on embedded Jetson TX2 [J]. Journal of Computer Applications, 2023, 43(1): 98-103.
[9]	Jianzhuang LIN, Wenzhong YANG, Sixiang TAN, Lexin ZHOU, Danni CHEN. Fusing filter enhancement and reverse attention network for polyp segmentation [J]. Journal of Computer Applications, 2023, 43(1): 265-272.
[10]	Kewen LI, Jiantao YANG, Zongchao HUANG. Improved YOLOv3 target detection based on boundary limit point features [J]. Journal of Computer Applications, 2023, 43(1): 81-87.
[11]	Yuhang WANG, Yongxia ZHOU, Liangwu WU. Pooling algorithm based on Gaussian function [J]. Journal of Computer Applications, 2022, 42(9): 2800-2806.
[12]	Zhihong XI, Jiaxu WEN. Indoor dynamic scene localization and mapping based on target detection [J]. Journal of Computer Applications, 2022, 42(9): 2853-2857.
[13]	Yuefeng LIU, Xiaoyan ZHANG, Wei GUO, Haodong BIAN, Yingjie HE. Remaining useful life prediction method of aero-engine based on optimized hybrid model [J]. Journal of Computer Applications, 2022, 42(9): 2960-2968.
[14]	Hongjun HENG, Tianbao XU. Attention sentiment analysis model based on multi-scale convolution and gating mechanism [J]. Journal of Computer Applications, 2022, 42(9): 2674-2679.
[15]	Zanxia QIANG, Xianfu BAO. Residual attention deraining network based on convolutional long short-term memory [J]. Journal of Computer Applications, 2022, 42(9): 2858-2864.

模型	耗时	模型	耗时
3D-FCN	0.22	3D-UNet	0.25
3D-SegNet	0.26	3D-CSA-UNet	0.24

模型	耗时	模型	耗时
3D-FCN	0.22	3D-UNet	0.25
3D-SegNet	0.26	3D-CSA-UNet	0.24