基于桥梁裂缝识别模型的桥梁裂缝病害识别方法

doi:10.13229/j.cnki.jdxbgxb.20210860

摘要/Abstract

摘要：

为实现高效、轻量化、无接触的桥梁裂缝病害识别，提出了一种基于桥梁裂缝识别模型（Bridge crack extraction model，BCEM）的桥梁裂缝识别网络。该网络将深度学习与传统图像处理方法相结合，首先，预处理裂缝图像，增强裂缝信息表达；之后，采用滑窗法将裂缝切分为面元图像，针对面元图像特性，采用改进的BC-MobileNet轻量化模型对裂缝面元进行分类；最后，识别误检与漏检面元，实现桥梁裂缝准确识别。通过与目标检测、模式识别等不同裂缝识别方式进行比较，结果表明：BCEM在各项实验指标上均有提升，证明了本文提出识别网络对桥梁裂缝识别的有效性。

关键词: 计算机应用, 深度学习, 裂缝检测, 图像处理, 卷积神经网络

Abstract:

To realize high-efficiency， light-weight， and non-contact bridge crack disease identification， a bridge crack identification network based on the （bridge crack extraction model，BCEM） is proposed. The network combines deep learning with traditional image processing methods. First， the crack image is preprocessed to enhance the expression of crack information. Then the sliding window method is used to divide the crack into patches. According to the characteristics of the patches， the improved lightweight model named BC-MobileNet is used to classify crack features. Finally， misdetected and undetected cracks are identified to achieve accurate identification of bridge cracks. Compared with different crack identification methods such as target detection and pattern recognition， the results show that the BCEM has improved in various experimental indicators， which proves the effectiveness of this network for bridge crack identification.

Key words: computer application, deep learning, crack detection, image processing, convolutional neural network

中图分类号:

TP391.41

张振海,季坤,党建武. 基于桥梁裂缝识别模型的桥梁裂缝病害识别方法[J]. 吉林大学学报(工学版), 2023, 53(5): 1418-1426.

Zhen-hai ZHANG,Kun JI,Jian-wu DANG. Crack identification method for bridge based on BCEM model[J]. Journal of Jilin University(Engineering and Technology Edition), 2023, 53(5): 1418-1426.

图/表 15

图1

图2

图3

图4

图5

图6

图7

图8

图9

表1

图10

表2

表3

图11

表4

参考文献 26

1	《中国公路学报》编辑部. 中国桥梁工程学术研究综述·2021[J]. 中国公路学报, 2021, 34(2): 1-97.
	Editorial Department of China Journal of Highway and Transport. A summary of academic research on chinese bridge engineering·2021[J]. Journal of China Highway, 2021, 34(2): 1-97.
2	Song B, Wei N. Statistics properties of asphalt pavement images for cracks detection[J]. Journal of Information & Computational Science, 2013, 10(9): 2833-2843.
3	Zou Q, Zhang Z, Li Q Q, et al. Deepcrack: learning hierarchical convolutional features for crack detection[J]. IEEE Trans Image Processing, 2019, 28(3): 1498-1512.
4	肖明尧, 李雄飞, 张小利, 等.基于多尺度的区域生长的图像分割算法[J].吉林大学学报: 工学版, 2017, 47(5): 1591-1597.
	Xiao Ming-yao, Li Xiong-fei, Zhang Xiao-li, et al. Image segmentation algorithm based on multi-scale region growth[J]. Journal of Jilin University(Engineering and Technology Edition), 2017, 47(5): 1591-1597.
5	Qu Z, Chen S Q, Liu Y Q, et al. Linear seam elimination of tunnel crack images based on statistical specific pixels ratio and adaptive fragmented segmentation[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 21: 3599-3607.
6	谢志江, 吕波, 刘琴, 等.旋转不变性图像模板匹配快速算法[J].吉林大学学报: 工学版, 2013, 43(3): 711-717.
	Xie Zhi-jiang, Bo LYU, Liu Qin, et al. A fast algorithm for image template matching with rotation invariance[J]. Journal of Jilin University(Engineering and Technology Edition), 2013, 43(3): 711-717.
7	Zhao B, Dai M, Li P, et al. Defect detection method for electric multiple units key components based on deep learning[J]. IEEE Access, 2020, 8: 136808-136818.
8	李良福, 马卫飞, 李丽, 等.基于深度学习的桥梁裂缝检测算法研究[J].自动化学报, 2019, 45(9): 1727-1742.
	Li Liang-fu, Ma Wei-fei, Li Li, et al. Research on bridge crack detection algorithm based on deep learning[J]. Automatica Sinica, 2019, 45(9): 1727-1742.
9	Cha Y J, Choi W, Oral B. Deep learning-based crack damage detection using convolutional neural networks[J]. Computer-Aided Civil and Infrastructure Engineering, 2017, 32(5): 361-378.
10	He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition[C]∥IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, USA, 2016: 770-778.
11	Rahul K, Ankur G, Harkirat S A, et al. CBSN: comparative measures of normalization techniques for brain tumor segmentation using SRCNet[J]. Multimedia Tools and Applications, 2022, 81(10): 13203-13235.
12	曹锦纲, 杨国田, 杨锡运.基于注意力机制的深度学习路面裂缝检测[J].计算机辅助设计与图形学学报, 2020, 32(8): 1324-1333.
	Cao Jin-gang, Yang Guo-tian, Yang Xi-yun. Deep learning pavement crack detection based on attention mechanism[J]. Journal of Computer Aided Design and Graphics, 2020, 32(8): 1324-1333.
13	黄宏伟, 李庆桐.基于深度学习的盾构隧道渗漏水病害图像识别[J].岩石力学与工程学报, 2017, 36(12): 2861-2871.
	Huang Hong-wei, Li Qing-tong. Image recognition of water leakage in shield tunnels based on deep learning[J]. Chinese Journal of Rock Mechanics and Engineering, 2017, 36(12): 2861-2871.
14	Simonyan S, Zisserman A. Very deep convolutional networks for large-scale image recognition[J/OL]. [2015-04-10].
15	Islam M M M, Kim JM. Vision-based autonomous crack detection of concrete structures using a fully convolutional encoder-decoder network[J]. Sensors(Basel), 2019(19): 19194251.
16	薛亚东, 李宜城.基于深度学习的盾构隧道衬砌病害识别方法[J].湖南大学学报: 自然科学版, 2018, 45(3): 100-109.
	Xue Ya-dong. Li Yi-cheng. A method of disease recognition forshield tunnel lining based on deep learning[J]. Joumal of Hunan University(Natural Sciences), 2018, 45(3): 100-109.
17	Sandler M, Howard A, Zhu M, et al. MobileNetV2: inverted residuals and linear bottlenecks[C]∥IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 4510-4520.
18	Redmon J, Farhadi A. YOLOv3: an incremental improvement[J/OL]. [2018-04-08].
19	蔡逢煌, 张岳鑫, 黄捷.基于YOLOv3与注意力机制的桥梁表面裂痕检测算法[J].模式识别与人工智能, 2020, 33(10): 926-933.
	Cai Feng-huang, Zhang Yue-xin, Huang Jie. Bridge surface crack detection algorithm based on YOLOv3 and attention mechanism[J]. Pattern Recognition and Artificial Intelligence, 2020, 33(10): 926-933.
20	Saouma V E, Barton C C, Gamaleldin N A. Fractal characterization of fracture surfaces in concrete[J]. Engineering Fracture Mechanics, 1990, 35(1-3): 47-53.
21	Xu F Y, Jiang Q S. Dynamic obstacle surmounting analysis of a bilateral-wheeled cable-climbing robot for cable-stayed bridges[J]. The Industrial Robot, 2019, 46(3): 431-443.
22	李超飞.桥梁病害检测的无人机地面站软件设计与实现[D].西安: 长安大学电子与控制工程学院, 2020.
	Li Chao-fei. Design and implementation of UAV ground station software for bridge disease detection [D]. Xi'an: School of Electronics and Control Engineering, Chang'an University, 2020.
23	Howard A G, Zhu M, Chen B, et al. MobileNets: efficient convolutional neural networks for mobile vision applications[J/OL]. [2017-04-17].
24	Shi Y, Cui L M, Qi Z Q, et al. Automatic road crack detection using random structured forests[J]. IEEE Ttransactions on Intelligent Transportation Systems, 2016, 17(12): 3434-3445.
25	Billah U H, La H M, Tavakkoli A. Deep learning-based feature silencing for accurate concrete crack detection[J]. Sensors, 2020, 20(16): 20164403.
26	Jiang W B, Liu M, Peng Y N, et al. HDCB-Net: a neural network with the hybrid dilated convolution for pixel-level crack detection on concrete bridges[J]. IEEE Transactions on Industrial Informatics, 2021, 17(8): 5485-5494.

相关文章 15

[1]	张则强,梁巍,谢梦柯,郑红斌. 混流双边拆卸线平衡问题的精英差分进化算法[J]. 吉林大学学报(工学版), 2023, 53(5): 1297-1304.
[2]	于鹏,朴燕. 基于多尺度特征的行人重识别属性提取新方法[J]. 吉林大学学报(工学版), 2023, 53(4): 1155-1162.
[3]	吴飞,农皓业,马晨浩. 基于粒子群优化算法⁃长短时记忆模型的刀具磨损预测方法[J]. 吉林大学学报(工学版), 2023, 53(4): 989-997.
[4]	姜宇,潘家铮,陈何淮,符凌智,齐红. 基于分割方法的繁体中文报纸文本检测[J]. 吉林大学学报(工学版), 2023, 53(4): 1146-1154.
[5]	何科,丁海涛,赖宣淇,许男,郭孔辉. 基于Transformer的轮式里程计误差预测模型[J]. 吉林大学学报(工学版), 2023, 53(3): 653-662.
[6]	吴振宇,刘小飞,王义普. 基于DKRRT^*-APF算法的无人系统轨迹规划[J]. 吉林大学学报(工学版), 2023, 53(3): 781-791.
[7]	陶博,颜伏伍,尹智帅,武冬梅. 基于高精度地图增强的三维目标检测算法[J]. 吉林大学学报(工学版), 2023, 53(3): 802-809.
[8]	薛珊,张亚亮,吕琼莹,曹国华. 复杂背景下的反无人机系统目标检测算法[J]. 吉林大学学报(工学版), 2023, 53(3): 891-901.
[9]	潘弘洋,刘昭,杨波,孙庚,刘衍珩. 基于新一代通信技术的无人机系统群体智能方法综述[J]. 吉林大学学报(工学版), 2023, 53(3): 629-642.
[10]	何颖,樊俊松,王巍,孙庚,刘衍珩. 无人机空地安全通信与航迹规划的多目标联合优化方法[J]. 吉林大学学报(工学版), 2023, 53(3): 913-922.
[11]	郭鹏,赵文超,雷坤. 基于改进Jaya算法的双资源约束柔性作业车间调度[J]. 吉林大学学报(工学版), 2023, 53(2): 480-487.
[12]	刘近贞,高国辉,熊慧. 用于脑组织分割的多尺度注意网络[J]. 吉林大学学报(工学版), 2023, 53(2): 576-583.
[13]	祁贤雨,王巍,王琳,赵玉飞,董彦鹏. 基于物体语义栅格地图的语义拓扑地图构建方法[J]. 吉林大学学报(工学版), 2023, 53(2): 569-575.
[14]	时小虎,吴佳琦,吴春国,程石,翁小辉,常志勇. 基于残差网络的弯道增强车道线检测方法[J]. 吉林大学学报(工学版), 2023, 53(2): 584-592.
[15]	高金武,贾志桓,王向阳,邢浩. 基于PSO-LSTM的质子交换膜燃料电池退化趋势预测[J]. 吉林大学学报(工学版), 2022, 52(9): 2192-2202.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

测试集面元总数	预处理	模型正确识别数	准确率/%
500	否	398	79.6
500	传统方法	389	77.8
500	本文方法	463	92.6

主干网络	测试集面元数	面元规格/像素	识别准确率/%	识别召回率/%
VGG-16	500	16×16	16.7	62.4
MobileNetV2	500	16×16	67.2	73.6
MobileNetV2-A	500	8×8	40.7	52.8
MobileNetV2-B	500	32×32	76.8	83.0
MobileNetV2-C	500	16×16	71.8	85.3
MobileNetV2-D	500	16×16	75.0	81.8
BC-MobileNet	500	16×16	92.6	94.1

模型	测试集面元数	平均识别IoU	AP	识别速度/（s·面元^-1）	模型大小/Mbit
BCEM	500	0.98	0.9555	1.762	19.9
Faster-Rcnn	500	0.87	0.9073	1.152	127
YoLoV3	500	0.82	0.9019	0.089	69

模型	测试集图片数	pr/%	re/%	模型大小/Mbit
Otsu	100	15.7	16.3	-
最大熵分割	100	39.7	52.4	-
漫水填充	100	7.8	7.5	-
FCN	100	72.2	75.2	39
U-Net	100	81.7	83.4	28
ANet-FSM^［25］	100	88.46	87.0	40
HDCB-Net^［26］	100	91.44	86.35	64
BCEM	100	97.5	98.7	19.9