«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《建筑科学与工程学报》[ISSN:1673-2049/CN:61-1442/TU]

卷:

39卷

期数:

2022年04期

页码:

127-136

栏目:

出版日期:

2022-07-12

文章信息/Info

Title:

Research on Classification and Recognition of Concrete Structure Diseases Based on Residual Network

文章编号:

1673-2049(2022)04-0127-10

作者:

丁 威^1,2,3,马亥波¹,舒江鹏¹,NIZHEGORODTSEV Denis V⁴,叶建龙⁵

(1. 浙江大学 建筑工程学院,浙江 杭州 310058; 2. 浙江大学 平衡建筑研究中心,浙江 杭州 310058; 3. 浙江大学建筑设计研究院有限公司,浙江 杭州 310058; 4. 圣彼得堡国立建筑工程大学 土木工程学院,圣彼得堡 190005; 5. 浙江数智交院科技股份有限公司,浙江 杭州 310058)

Author(s):

DING Wei^1,2,3, MA Hai-bo¹, SHU Jiang-peng¹, NIZHEGORODTSEV Denis V⁴, YE Jian-long⁵

(1.College of Civil Engineering and Architecture,Zhejiang University,Hangzhou 310058,Zhejiang,China;2.Center for Balance Architecture,Zhejiang University,Hangzhou 310058,Zhejiang,China;3.The Architecture Design&Research Institution of Zhejiang University Co.,Ltd,Hangzhou 310058,Zhejiang,China;4.Faculty of Civil Engineering,Saint-Petersburg State University of Architecture and Civil Engineering,Saint-Petersburg 190005,Russia;

关键词:

混凝土结构;  病害分类识别;  残差网络;  多属性标注;  迁移学习

Keywords:

concrete structure;  disease classification and recognition;  residual network;  multiattribute label;  transfer learning

分类号:

TU317

DOI:

10.19815/j.jace.2021.10112

文献标志码:

A

摘要:

针对混凝土结构病害识别类型单一、精度较低的现状,提出了基于残差网络和迁移学习的病害分类识别方法,通过构建多属性病害数据集,利用迁移学习优化残差网络模型,提出混凝土结构健康状态识别的多个任务。首先收集混凝土结构的病害状态图像,依次通过数据清洗、尺寸均一化、数据扩增和多人投票标注,最终得到包含6 680张图像的混凝土结构病害多属性数据集,并依据不同标注属性进行了相应训练集、验证集和测试集的划分; 然后利用迁移学习对预训练的ResNet-34网络前3个部分进行参数冻结,后续2个部分的参数进行重新训练,并在模型末端添加新的参数,基于已构建的数据集进行训练; 最后在提出的构件类别检测、剥落检测、病害检测和病害类别检测任务中,分别获得84.88%、98.56%、97.18%和85.34%的F1分数。结果表明:通过构建多属性标注的混凝土结构病害数据集训练深度学习模型,可较好地实现多场景特征下的病害识别效果; 采用迁移学习技术可从开源数据中获取较好的特征提取效果; 改进的ResNet-34网络可克服网络退化问题,并针对混凝土结构病害识别的多个任务获得较好的效果; 相对于单一的混凝土结构病害识别,进行病害部位、程度、多类别的系统性检测,可为结构状态评估提供详细信息,更贴合工程实际需要。

Abstract:

In view of the single type and low accuracy of concrete structure disease recognition, a disease classification and recognition method based on residual network and transfer learning was proposed. By establishing a multiattribute disease dataset and using transfer learning to improve the residual network, the multiple tasks for identifying the health status of concrete structures were proposed. First of all, the original images of disease state of concrete structure were collected and sequentially underwent data cleaning, size normalization, data augmentation and multi-person voting to label. Then, a multiattribute dataset of concrete structure diseases containing 6 680 images was obtained, and corresponding training set, validation set and testing set were divided according to different label attributes. Sequentially, the first three parts of parameters of the pre-trained ResNet-34 were frozen by transfer learning. The parameters of subsequent two parts were retrained, and new parameters were added at the end of the model and training was conducted based on the established dataset. Finally, the F1 score of 84.88%, 98.56%, 97.18% and 85.34% were obtained in the proposed tasks of component type detection, spalling detection, disease detection and disease type detection, respectively. The results show that by building a multiattribute labeled dataset of concrete structure diseases to train a deep learning model, the disease recognition under multi-scene conditions can be better achieved. Better feature extraction results can be obtained from open source data by using transfer learning.The improved ResNet-34 can overcome the problem of network degradation, and achieve better results for multiple tasks of concrete structure disease identification. Compared with the single disease identification of concrete structure, systematic detection of disease location, degree, and multiple categories can provide detailed information for structural state assessment, which is more suitable for the actual needs of the project.

参考文献/References:

[1] 《中国公路学报》编辑部.中国桥梁工程学术研究综述·2014[J].中国公路学报,2014,27(5):1-96.
Editorial Department of China Journal of Highway and Transport.Review on China's Bridge Engineering Research:2014[J].China Journal of Highway and Transport,2014,27(5):1-96.
[2]赵士良,韩万水,鲁永飞,等.重载交通条件下装配式RC板桥抗裂性分析[J].建筑科学与工程学报,2015,32(4):73-79.
ZHAO Shi-liang,HAN Wan-shui,LU Yong-fei,et al.Crack Resistance Analysis on Prefabricated RC Slab Bridge Under Heavy Traffic[J].Journal of Architecture and Civil Engineering,2015,32(4):73-79.
[3]ABDEL-QADER I,ABUDAYYEH O,KELLY M E.Analysis of Edge-detection Techniques for Crack Identification in Bridges[J].Journal of Computing in Civil Engineering,2003,17(4):255-263.
[4]YAMAGUCHI T,NAKAMURA S,SAEGUSA R,et al.Image-based Crack Detection for Real Concrete Surfaces[J].IEEJ Transactions on Electrical and Electronic Engineering,2008,3(1):128-135.
[5]刘宇飞.基于模型修正与图像处理的多尺度结构损伤识别[D].北京:清华大学,2015.
LIU Yu-fei.Multi-scale Structural Damage Assessment Based on Model Updating and Image Processing[D].Beijing:Tsinghua University,2015.
[6]于华洋,马 涛,王大为,等.中国路面工程学术研究综述·2020[J].中国公路学报,2020,33(10):1-66.
YU Hua-yang,MA Tao,WANG Da-wei,et al.Review on China's Pavement Engineering Research·2020[J].China Journal of Highway and Transport,2020,33(10):1-66.
[7]ZHANG L,YANG F,DANIEL ZHANG Y,et al.Road Crack Detection Using Deep Convolutional Neural Network[C]//IEEE.2016 IEEE International Conference on Image Processing.Phoenix:IEEE,2016:3708-3712.
[8]CHA Y J,CHOI W,BUYUKOZTURK O.Deep Learning-based Crack Damage Detection Using Convolutional Neural Networks[J].Computer-aided Civil and Infrastructure Engineering,2017,32(5):361-378.
[9]CHA Y J,CHOI W,SUH G,et al.Autonomous Structural Visual Inspection Using Region-based Deep Learning for Detecting Multiple Damage Types[J].Computer-aided Civil and Infrastructure Engineering,2018,33(9):731-747.
[10]汤一平,胡克钢,袁公萍.基于全景图像CNN的隧道病害自动识别方法[J].计算机科学,2017,44(增2):207-211,250.
TANG Yi-ping,HU Ke-gang,YUAN Gong-ping.Automatic Recognition Method of Tunnel Disease Based on Convolutional Neural Network for Panoramic Images[J].Computer Science,2017,44(S2):207-211,250.
[11]沙爱民,蔡若楠,高 杰,等.基于级联卷积神经网络的公路路基病害识别[J].长安大学学报(自然科学版),2019,39(2):1-9.
SHA Ai-min,CAI Ruo-nan,GAO Jie,et al.Subgrade Distresses Recognition Based on Convolutional Neural Network[J].Journal of Chang'an University(Natural Science Edition),2019,39(2):1-9.
[12]沙爱民,童 峥,高 杰.基于卷积神经网络的路表病害识别与测量[J].中国公路学报,2018,31(1):1-10.
SHA Ai-min,TONG Zheng,GAO Jie.Recognition and Measurement of Pavement Disasters Based on Convolutional Neural Networks[J].China Journal of Highway and Transport,2018,31(1):1-10.
[13]XU Y,BAO Y Q,CHEN J H,et al.Surface Fatigue Crack Identification in Steel Box Girder of Bridges by a Deep Fusion Convolutional Neural Network Based on Consumer-grade Camera Images[J].Structural Health Monitoring,2019,18(3):653-674.
[14]LI S Y,ZHAO X F,ZHOU G Y.Automatic Pixel-level Multiple Damage Detection of Concrete Structure Using Fully Convolutional Network[J].Computer-aided Civil and Infrastructure Engineering,2019,34(7):616-634.
[15]朱劲松,李 欢,王世芳.基于卷积神经网络和迁移学习的钢桥病害识别[J].长安大学学报(自然科学版),2021,41(3):52-63.
ZHU Jin-song,LI Huan,WANG Shi-fang.Defect Recognition for Steel Bridge Based on Convolutional Neural Network and Transfer Learning[J].Journal of Chang'an University(Natural Science Edition),2021,41(3):52-63.
[16]HE K M,ZHANG X Y,REN S Q,et al.Deep Residual Learning for Image Recognition[C]//IEEE.2016 IEEE Conference on Computer Vision and Pattern Recognition.Las Vegas:IEEE,2016:770-778.
[17]DENG J,DONG W,SOCHER R,et al.ImageNet:A Large-scale Hierarchical Image Database[C]//IEEE.2009 IEEE Conference on Computer Vision and Pattern Recognition.Miami:IEEE,2009:248-255.
[18]GAO Y Q,MOSALAM K M.PEER Hub ImageNet:A Large-scale Multiattribute Benchmark Data Set of Structural Images[J].Journal of Structural Engineering,2020,146(10):04020198.
[19]ZHANG H N,DONG B,FENG B Q,et al.Classification of Financial Tickets Using Weakly Supervised Fine-grained Networks[J].IEEE Access,2020,8:129469-129477.
[20]PASZKE A,GROSS S,MASSA F,et al.Pytorch:An Imperative Style,High-performance Deep Learning Library[J].Advances in Neural Information Processing Systems,2019,32:8026-8037.
[21]BENGIO Y.Neural Networks:Tricks of the Trade[M].2nd ed.Berlin:Springer,2012.

相似文献/References:

[1]曹 杨,孙千伟,宫文军,等.新型装配式混凝土框架型钢节点试验[J].建筑科学与工程学报,2016,33(02):15.
　CAO Yang,SUN Qian-wei,GONG Wen-jun,et al.Experiment on New Section Steel Joints for Prefabricated Concrete Frame[J].Journal of Architecture and Civil Engineering,2016,33(04):15.
[2]陆春华,袁思奇.基于时变可靠度的锈蚀混凝土结构全寿命成本模型[J].建筑科学与工程学报,2017,34(02):71.
　LU Chun-hua,YUAN Si-qi.Life-cycle Cost Model of Corroded Concrete Structures Based on Time-varying Reliability[J].Journal of Architecture and Civil Engineering,2017,34(04):71.
[3]何化南,冯 叶,张冠华,等.考虑表面氯离子质量分数的沿海混凝土桥梁氯离子扩散修正模型[J].建筑科学与工程学报,2017,34(04):57.
　HE Hua-nan,FENG Ye,ZHANG Guan-hua,et al.Revised Chloride Ion Diffusion Models of Concrete Bridges Near Coastal Areas Considering Surface Chloride Ion Concentration[J].Journal of Architecture and Civil Engineering,2017,34(04):57.
[4]陈曦,梁剑青.混凝土结构周期比分析[J].建筑科学与工程学报,2009,26(03):117.
　CHEN Xi,LIANG Jian-qing.Analysis of Period Ratio of Concrete Structure[J].Journal of Architecture and Civil Engineering,2009,26(04):117.
[5]马昆林,谢友均,龙广成.氯盐环境下桥梁混凝土结构的腐蚀行为及破坏机理[J].建筑科学与工程学报,2008,25(03):32.
　MA Kun-lin,XIE You-jun,LONG Guang-cheng.Corrosion Behavior and Destructive Mechanism of Bridge Concrete Structure Under Chloride Salt Environment[J].Journal of Architecture and Civil Engineering,2008,25(04):32.
[6]肖建庄,陈立浩,叶建军,等.混凝土结构拆除技术与绿色化发展[J].建筑科学与工程学报,2019,36(05):1.
　XIAO Jian-zhuang,CHEN Li-hao,YE Jian-jun,et al.Technology and Green Development of Demolition for Concrete Structures[J].Journal of Architecture and Civil Engineering,2019,36(04):1.
[7]陈 琳,屈文俊,朱 鹏,等.混凝土结构等耐久性设计方法[J].建筑科学与工程学报,2020,37(02):81.[doi:10.19815/j.jace.2019.03059]
　CHEN Lin,QU Wen-jun,ZHU Peng,et al.Equal Durability Design Method for Concrete Structure[J].Journal of Architecture and Civil Engineering,2020,37(04):81.[doi:10.19815/j.jace.2019.03059]
[8]任 伟,李敬泉,李晓路.带翼缘截面混凝土受压区等效应力图系数研究[J].建筑科学与工程学报,2022,39(06):87.[doi:10.19815/j.jace.2021.09078]
　REN Wei,LI Jing-quan,LI Xiao-lu.Study on Equivalent Stress Diagram Coefficient of Concrete Compression Zone with Flange Section[J].Journal of Architecture and Civil Engineering,2022,39(04):87.[doi:10.19815/j.jace.2021.09078]

备注/Memo

备注/Memo:

收稿日期:2021-10-29
基金项目:国家自然科学基金青年科学基金项目(52108179); 浙江省科学技术厅重点研发计划项目(2021C01106)
作者简介:丁 威(1994-),男,湖北鄂州人,工学博士研究生,E-mail:weiding1029@zju.edu.cn。
通信作者:舒江鹏(1987-),男,浙江杭州人,教授,博士研究生导师,工学博士,E-mail:jpeshu@zju.edu.cn。

常用功能

更新日期/Last Update: 2022-07-10