[1]郎洪,温添,陆键,等.基于深度学习的三维路面裂缝类病害检测方法[J].东南大学学报(自然科学版),2021,(1):53-60.[doi:10.3969/j.issn.1001-0505.2021.01.008]
　Lang Hong,Wen Tian,Lu Jian,et al.3D pavement crack detection method based on deep learning[J].Journal of Southeast University (Natural Science Edition),2021,(1):53-60.[doi:10.3969/j.issn.1001-0505.2021.01.008]
点击复制

基于深度学习的三维路面裂缝类病害检测方法()
分享到：

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
期数:
2021年第1期
页码:
53-60
栏目:
交通运输工程
出版日期:
2021-01-20

文章信息/Info

Title:
3D pavement crack detection method based on deep learning
作者:
郎洪1; 温添1; 陆键 1; 丁朔1; 陈圣迪2
1 同济大学道路与交通工程教育部重点实验室, 上海 201804; 2 上海海事大学交通运输学院, 上海 201306
Author(s):
Lang Hong1;  Wen Tian1;  Lu Jian1;  Ding Shuo1;  Chen Shengdi2
1 Key Laboratory of Road and Traffic Engineering of Ministry of Education, Tongji University, Shanghai 201804, China
2 College of Transportation Engineering, Shanghai Maritime University, Shanghai 201306, China
关键词:
道路工程;  裂缝检测;  三维图像;  卷积神经网络;  高程检查
Keywords:
road engineering;  crack detection;  3D image;  convolutional neural networks;  depth-checking
分类号:
U416.2
DOI:
10.3969/j.issn.1001-0505.2021.01.008
摘要:
为了在路面三维图像的基础上快速、准确、完整地识别裂缝,提出一种基于深度学习的路面裂缝类病害自动检测方法.首先,以子块图像为处理单元,将三维图像划分为裂缝面元和背景面元,其中背景面元包含了路面标线、不同纹理和桥接缝等复杂场景.根据对面元图像的分析,提出一种基于卷积神经网络的PCCNet分类模型,用于路面背景面元和路面裂缝面元的自动识别.然后,为了进一步提取裂缝面元内裂缝的完整轮廓,考虑路面三维图像中裂缝像素级邻域特征,利用PCCNet模型结合裂缝高程检查方法对路面裂缝进行检测.研究结果表明:通过训练集4 300张高精度三维图像的训练,模型在3 850次迭代之后出现过拟合,且此时PCCNet模型在验证集上的总体F值达到最大,为92.9%;将PCCNet模型结合裂缝高程检查方法应用在测试集的200张三维图像上,方法准确率、召回率和F值分别为87.8%、90.1%和88.9%.与改进Canny方法和种子识别方法对比,所提出的方法在抑制噪声和检测细小裂纹方面具有更强的鲁棒性.
Abstract:
In order to detect cracks quickly, accurately and completely on the basis of three-dimensional(3D)pavement images, an automatic detection method of pavement cracks based on the deep learning model is proposed. Firstly, taking the sub-block image as the processing unit, the 3D image is divided into crack patches and background patches, in which the background patch contains complex scenes such as pavement marking, different textures and bridge joints. A pavement crack classification network(PCCNet)based on the convolutional neural network(CNN)is proposed for the automatic recognition of pavement background patch and pavement crack patch. Then, in order to further extract the complete contour of cracks, considering the pixel-level neighborhood features of cracks in 3D pavement images, the PCCNet combined with the crack depth-checking method is used to detect pavement cracks. The results show that through the training of 4 300 high-precision 3D images in the training set, the model is over fitted after 3 850 iterations, and the overall F-value of the PCCNet on the verification set reaches the maximum, which is 92.9%; the PCCNet combined with the depth-checking method is applied to 200 3D images in the testing set, and the accuracy rate, recall rate and F-value are 87.8%, 90.1% and 88.9%, respectively. Compared with the improved Canny algorithm and the seed recognition method, the proposed method has stronger robustness in suppressing noise and detecting small cracks.

参考文献/References:

[1] Serigos P A, Prozzi J A, Smit A D F, et al. Evaluation of 3D automated systems for the measurement of pavement surface cracking[J].Journal of Transportation Engineering, 2016, 142(6): 05016003. DOI:10.1061/(ASCE)TE.1943-5436.0000841.
[2] 徐志刚, 车艳丽, 李金龙, 等. 路面破损图像自动处理技术研究进展[J]. 交通运输工程学报, 2019, 19(1): 172-190.
　　Xu Z G, Che Y L, Li J L, et al. Research progress on automatic image processing technology for pavement distress[J]. Journal of Traffic and Transportation Engineering, 2019, 19(1): 172-190.(in Chinese)
[3] Laurent J, Hébert J F, Lefebvre D, et al. Using 3D laser profiling sensors for the automated measurement of road surface conditions[M]//7th RILEM International Conference on Cracking in Pavements. Dordrecht: Springer Netherlands, 2012: 157-167. DOI:10.1007/978-94-007-4566-7_16.
[4] Kaushik R, Xiao J, Morris W, et al. 3D laser scan registration of dual-robot system using vision [C]// International Conference on Intelligent Robots and Systems. Louis, MO, USA, 2009: 4148-4153. DOI: 10.1109/IROS.2009.5354773.
[5] Pollefeys M, Sinha S N, Zach C, et al. Discovering and exploiting 3D symmetries in structure from motion [C]// IEEE Conference on Computer Vision and Pattern Recognition. Providence, RI, USA, 2012: 1514-1521. DOI: 10.1109/CVPR.2012.6247841.
[6] Scherer S. Focus variation—a new technology for high resolution optical 3D surface metrology [J]. Journal of Mechanical Engineering, 2011, 57(3):245-256. DOI: 10.5545/sv-jme.2010.175.
[7] Wöhler C. Triangulation-based approaches to three-dimensional scene reconstruction [M]// 3D Computer Vision. London: Springer London, 2013: 3-87.
[8] Fujimoto J G, Pitris C, Boppart S A, et al. Optical coherence tomography: An emerging technology for biomedical imaging and optical biopsy[J].Neoplasia, 2000, 2(1/2): 9-25. DOI:10.1038/sj.neo.7900071.
[9] Mathavan S, Kamal K, Rahman M. A review of three-dimensional imaging technologies for pavement distress detection and measurements[J].IEEE Transactions on Intelligent Transportation Systems, 2015, 16(5): 2353-2362. DOI:10.1109/tits.2015.2428655.
[10] Jahanshahi M R, Jazizadeh F, Masri S F, et al. Unsupervised approach for autonomous pavement-defect detection and quantification using an inexpensive depth sensor[J].Journal of Computing in Civil Engineering, 2013, 27(6): 743-754. DOI:10.1061/(asce)cp.1943-5487.0000245.
[11] Zhang A, Wang K C P, Ji R, et al. Efficient system of cracking-detection algorithms with 1-mm 3D-surface models and performance measures[J]. Journal of Computing in Civil Engineering, 2016, 30(6): 04016020. DOI:10.1061/(asce)cp.1943-5487.0000581.
[12] Sollazzo G, Wang K C P, Bosurgi G, et al. Hybrid procedure for automated detection of cracking with 3D pavement data[J]. Journal of Computing in Civil Engineering, 2016, 30(6): 04016032. DOI:10.1061/(asce)cp.1943-5487.0000597.
[13] Zhang A, Wang K C P, Ai C F. 3D shadow modeling for detection of descended patterns on 3D pavement surface[J].Journal of Computing in Civil Engineering, 2017, 31(4): 04017019. DOI:10.1061/(asce)cp.1943-5487.0000661.
[14] Zhang A, Wang K C P, Li B X, et al. Automated pixel-level pavement crack detection on 3D asphalt surfaces using a deep-learning network[J]. Computer-Aided Civil and Infrastructure Engineering, 2017, 32(10): 805-819. DOI:10.1111/mice.12297.
[15] Kavitha K, Sandhya B, Thirumala B. Evaluation of distance measures for feature based image registration using AlexNet[J]. International Journal of Advanced Computer Science and Applications, 2018, 9(10):284-290. DOI:10.14569/ijacsa.2018.091034.
[16] Bi N, Chen J H, Tan J. The handwritten Chinese character recognition uses convolutional neural networks with the GoogLeNet[J]. International Journal of Pattern Recognition and Artificial Intelligence, 2019, 33(11): 1940016. DOI:10.1142/s0218001419400160.
[17] Ren S Q, He K M, Girshick R, et al. Faster R-CNN: Towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149. DOI:10.1109/tpami.2016.2577031.
[18] Li B X, Wang K C P, Zhang A, et al. Automatic classification of pavement crack using deep convolutional neural network[J]. International Journal of Pavement Engineering, 2020, 21(4): 457-463. DOI:10.1080/10298436.2018.1485917.
[19] Zhang L, Yang F, Zhang D, et al. Road crack detection using deep convolutional neural network [C]// IEEE International Conference on Image Processing. Phoenix, AZ, USA, 2016: 3708-3712. DOI: 10.1109/ICIP.2016.7533052.
[20] Cha Y J, Choi W, Büyüköztürk O. Deep learning-based crack damage detection using convolutional neural networks[J].Computer-Aided Civil and Infrastructure Engineering, 2017, 32(5): 361-378. DOI:10.1111/mice.12263.
[21] 中华人民共和国交通运输部.公路技术状况评定标准: JTG 5210—2018 [S]. 北京: 人民交通出版社, 2018.
[22] 彭凯飞, 沈学举, 黄富瑜, 等. 基于二分查找的Radon变换直线检测改进算法[J]. 半导体光电, 2019, 40(3): 438-443. DOI: 10.16818/j.issn1001-5868.2019.03.028.
Peng K F, Shen X J, Huang F Y, et al. Radon transform algorithms in line detection based on binary search[J]. Semiconductor Optoelectronics, 2019, 40(3): 438-443. DOI:10.16818/j.issn1001-5868.2019.03.028. (in Chinese)
[23] 郎洪,陆键,陈圣迪,等. 考虑病害三维特征的沥青路面车辙异常检验方法 [J]. 东南大学学报(自然科学版),2020,50(3):454-462. DOI: 10.3969/j.issn. 1001-0505.2020.03.007.
Lang H, Lu J, Chen S D, et al. Asphalt pavement rutting anomaly inspection method considering 3D characteristics of distress [J]. Journal of Southeast University(Natural Science Edition), 2020, 50(3): 454-462. DOI:10.3969/j.issn. 1001-0505.2020.03.007. (in Chinese)
[24] Zhao X M, Wang W X, Wang L P. Parameter optimal determination for canny edge detection[J].The Imaging Science Journal, 2011, 59(6): 332-341. DOI:10.1179/136821910x12867873897517.

相似文献/References:

[1]魏建军,邢姣秀,付智.行车荷载引起桥梁振动对修复混凝土性能影响[J].东南大学学报(自然科学版),2010,40(5):1057.[doi:10.3969/j.issn.1001-0505.2010.05.033]
　Wei Jianjun,Xing Jiaoxiu,Fu Zhi.Effect of traffic load induced bridge vibrations on concrete tensile properties[J].Journal of Southeast University (Natural Science Edition),2010,40(1):1057.[doi:10.3969/j.issn.1001-0505.2010.05.033]
[2]倪富健,成晟,顾兴宇,等.路面结构的动态谱元分析[J].东南大学学报(自然科学版),2010,40(3):575.[doi:10.3969/j.issn.1001-0505.2010.03.027]
　Ni Fujian,Cheng Sheng,Gu Xingyu,et al.Spectral element analysis of dynamic pavement structure response[J].Journal of Southeast University (Natural Science Edition),2010,40(1):575.[doi:10.3969/j.issn.1001-0505.2010.03.027]
[3]马涛,张道义,黄晓明.SBS改性沥青抽提回收影响因素及改进方案[J].东南大学学报(自然科学版),2008,38(5):811.[doi:10.3969/j.issn.1001-0505.2008.05.014]
　Ma Tao,Zhang Daoyi,Huang Xiaoming.Influential factors and improvement of extraction and recovery of SBS modified asphalt[J].Journal of Southeast University (Natural Science Edition),2008,38(1):811.[doi:10.3969/j.issn.1001-0505.2008.05.014]
[4]黄宝涛,廖公云,张静芳.半刚性基层沥青路面层间接触临界状态值的计算方法[J].东南大学学报(自然科学版),2007,37(4):666.[doi:10.3969/j.issn.1001-0505.2007.04.024]
　Huang Baotao,Liao Gongyun,Zhang Jingfang.Analytical method of interlayer contact fettle in semi-rigid-base bituminous pavement[J].Journal of Southeast University (Natural Science Edition),2007,37(1):666.[doi:10.3969/j.issn.1001-0505.2007.04.024]
[5]陈俊,黄晓明.采用离散元方法评价集料的骨架结构[J].东南大学学报(自然科学版),2012,42(4):761.[doi:10.3969/j.issn.1001-0505.2012.04.035]
　Chen Jun,Huang Xiaoming.Evaluation of aggregate skeleton structure using the discrete element method[J].Journal of Southeast University (Natural Science Edition),2012,42(1):761.[doi:10.3969/j.issn.1001-0505.2012.04.035]
[6]孙志林,黄晓明.沥青路面线性疲劳损伤特性及形变规律[J].东南大学学报(自然科学版),2012,42(3):521.[doi:10.3969/j.issn.1001-0505.2012.03.025]
　Sun Zhilin,Huang Xiaoming.Linear fatigue damage characteristics and deformation law of asphalt pavement[J].Journal of Southeast University (Natural Science Edition),2012,42(1):521.[doi:10.3969/j.issn.1001-0505.2012.03.025]
[7]王艳,倪富健,李再新.水泥稳定碎石基层温缩性能试验及预估控制[J].东南大学学报(自然科学版),2008,38(2):260.[doi:10.3969/j.issn.1001-0505.2008.02.015]
　Wang Yan,Ni Fujian,Li Zaixin.Test and estimate control on temperature shrinkage performance of cement-treated macadam[J].Journal of Southeast University (Natural Science Edition),2008,38(1):260.[doi:10.3969/j.issn.1001-0505.2008.02.015]
[8]李志栋,黄晓明,岳学军.半刚性基层沥青路面非连续结构强迫振动声效[J].东南大学学报(自然科学版),2011,41(6):1277.[doi:10.3969/j.issn.1001-0505.2011.06.028]
　Li Zhidong,Huang Xiaoming,Yue Xuejun.Sound analysis of forced vibration feature of discontinuous structure in semi-rigid base asphalt pavement[J].Journal of Southeast University (Natural Science Edition),2011,41(1):1277.[doi:10.3969/j.issn.1001-0505.2011.06.028]
[9]孙璐,辛宪涛,任皎龙.纳米改性沥青混合料路用性能[J].东南大学学报(自然科学版),2013,43(4):873.[doi:10.3969/j.issn.1001-0505.2013.04.037]
　Sun Lu,Xin Xiantao,Ren Jiaolong.Pavement performance of nanomaterial modified asphalt mixture[J].Journal of Southeast University (Natural Science Edition),2013,43(1):873.[doi:10.3969/j.issn.1001-0505.2013.04.037]
[10]陈俊,黄晓明.路面加铺后旧沥青混合料的疲劳性能[J].东南大学学报(自然科学版),2008,38(3):516.[doi:10.3969/j.issn.1001-0505.2008.03.030]
　Chen Jun,Huang Xiaoming.Fatigue performance of old pavement asphalt mixtures after overlay[J].Journal of Southeast University (Natural Science Edition),2008,38(1):516.[doi:10.3969/j.issn.1001-0505.2008.03.030]

备注/Memo

备注/Memo:
收稿日期: 2020-08-04.
作者简介: 郎洪(1994—),男,博士生;陆键(联系人),男,教授,博士生导师,jianjohnlu@sina.com.
基金项目: 国家自然科学基金资助项目(71871165).
引用本文: 郎洪,温添,陆键,等.基于深度学习的三维路面裂缝类病害检测方法[J].东南大学学报(自然科学版),2021,51(1):53-60. DOI:10.3969/j.issn.1001-0505.2021.01.008.
更新日期/Last Update: 2021-01-20