|本期目录/Table of Contents|

[1]孙己龙,刘勇,路鑫,等.基于可变形卷积网络和YOLOv8的衬砌裂缝检测模型研究*[J].中国安全生产科学技术,2024,20(8):181-189.[doi:10.11731/j.issn.1673-193x.2024.08.024]
　SUN Jilong,LIU Yong,LU Xin,et al.Research on detection model of lining crack based on deformable convolutional network and YOLOv8[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2024,20(8):181-189.[doi:10.11731/j.issn.1673-193x.2024.08.024]
点击复制

基于可变形卷积网络和YOLOv8的衬砌裂缝检测模型研究*
分享到：

《中国安全生产科学技术》[ISSN:1673-193X/CN:11-5335/TB]

卷:
20
期数:
2024年8期
页码:
181-189
栏目:
职业安全卫生管理与技术
出版日期:
2024-08-30

文章信息/Info

Title:
Research on detection model of lining crack based on deformable convolutional network and YOLOv8
文章编号:
1673-193X(2024)-08-0181-09
作者:
孙己龙; 刘勇; 路鑫; 王志丰; 王亚琼; 侯小龙
（1.陕西省交通运输工程质量监测鉴定站，陕西 西安 710075；
2.长安大学 公路学院，陕西 西安 710064;
3.长安大学 材料科学与工程学院，陕西 西安 710064；
4.西安公路研究院有限公司，陕西 西安 710065)
Author(s):
SUN Jilong;  LIU Yong;  LU Xin;  WANG Zhifeng;  WANG Yaqiong;  HOU Xiaolong
(1.Shaanxi Provincial Transportation Engineering Quality Monitoring and Appraisal Station,Xi’an Shaanxi 710075,China;
2.School of Highway,Chang’an University,Xi’an Shaanxi 710064,China;
3.School of Materials Science and Engineering,Chang’an University,Xi’an Shaanxi 710064,China;
4.Xi’an Highway Research Institute Co.,Ltd.,Xi’an Shaanxi 710065,China)
关键词:
隧道工程; 结构安全; 可变形卷积网络; 衬砌裂缝; YOLOv8
Keywords:
tunnel engineering;  structural safety;  deformable convolutional network;  lining crack;  You Only Look Once v8 (YOLOv8)
分类号:
X951
DOI:
10.11731/j.issn.1673-193x.2024.08.024
文献标志码:
A
摘要:
为解决裂缝性状发育随机度高、标注框分辨率低、分布密集易重叠、目标相对小等因素引起的智能检测精度及效率差等问题，基于改进可变形卷积神经网络对YOLOv8骨干网络进行融合，提出1种能够适应隧道复杂场景的裂缝检测模型D-YOLO。模型首先对第3版可变形卷积网络(DCNv3)的空间聚合权重softmax归一化步骤进行去除以增强网络卷积效率，再利用新DCNv4对骨干网络C2f卷积模块进行融合以提升对网络图像中不同尺度裂缝性状及空间位置变化的细节感知能力，并采用自建裂缝数据集对SSD，Faster-RCNN，YOLOv5和YOLOv8 4种检测模型进行对比验证。研究结果表明：D-YOLO的F1分数为80.82%，mAP@0.5为86.90%，相较于SSD、Faster-RCNN、YOLOv5和YOLOv8都有所提升；D-YOLO的单张图像检测速度为20.36 ms，相较于各种对比模型分别加快37.06%、65.33%、45.22%和28.39%；同时，D-YOLO对衬砌裂缝图像特征关注范围有所增加。研究结果可为隧道运营期内衬砌安全检测提供新思路。
Abstract:
In order to solve the problems of poor intelligent detection accuracy and efficiency caused by the factors such as high randomness of crack characteristic development,low resolution of annotation box,dense distribution and easy overlap,and relatively small target,the YOLOv8 backbone network was fused based on the improved deformable convolutional neural network,and a crack detection model D-YOLO that can adapt to complex tunnel scenes was proposed.The normalization step of spatial aggregation weight softmax in the deformable convolutional network v3 (DCNv3) was removed to enhance the convolutional efficiency of network,and the new DCNv4 was used to fuse the C2f convolution module of backbone network to enhance the detail perception ability of different scale crack characteristics and spatial position change in the network images.The self-built crack dataset was used to compare and verify four detection models including SSD,Faster-RCNN,YOLOv5,and YOLOv8.The results show that the F1 score of D-YOLO is 80.82%,mAP@0.5 is 86.90%,and both of them are improved than those of SSD,Faster-RCNN,YOLOv5,and YOLOv8.The single image detection speed of D-YOLO is 20.36 ms,which is 37.06%,65.33%,45.22%,and 28.39% faster than those of various comparison models,respectively.Meanwhile,the attention range of image features of lining crack is increased through D-YOLO.The research results can provide new ideas for the safety detection of lining during tunnel operation.

参考文献/References:

［1］路鑫,刘勇,陈才,等.裂缝性状对隧道衬砌结构受力特征及安全性能影响研究［J/OL］.建筑科学与工程学报,1-14［2024-02-25］.http://kns.cnki.net/kcms/detail/61.1442.TU.20231128.1124.002.html.
［2］刘勇,王志丰,王亚琼,等.基于案例推理的隧道洞门智能设计方法研究［J］.铁道标准设计,2024,68(6):137-144. LIU Yong,WANG Zhifeng,WANG Yaqiong,et al.Investigation on intelligent design method of tunnel portal based on case-based reasoning ［J］.Railway Standard Design,2024,68(6):137-144.
［3］李宇辉,费瑞振,钱世嘉.考虑分岔角度和火源位置的地铁分岔隧道火灾临界风速研究［J］.中国安全生产科学技术,2024,20(1):100-106. LI Yuhui,FEI Ruizhen,QIAN Shijia.Study on critical wind velocity of subway bifurcated tunnel fire considering bifurcation angle and fire source location ［J］.Journal of Safety Science and Technology,2024,20(1):100-106.
［4］陈江,阳军生,曹能学,等.铁路隧道衬砌裂缝整治及衬砌补强加固设计研究［J］.中国安全生产科学技术,2014,10(9):134-139. CHEN Jiang,YANG Junsheng,CAO Nengxue,et al.Study on crack remediation and reinforcement of lining in railway tunnel ［J］.Journal of Safety Science and Technology,2014,10(9):134-139.
［5］张春海,李明雄,陈健,等.围岩松动及裂缝对隧道衬砌承载性能影响研究［J］.地下空间与工程学报,2022,18(增刊2):729-736. ZHANG Chunhai,LI Mingxiong,CHEN Jian,et al.Effect of rock loosening and concrete cracking on the performance of tunnel lining ［J］.Chinese Journal of Underground Space and Engineering,2022,18(Supplement 2):729-736.
［6］周中,张俊杰,鲁四平.基于改进YOLOv4的隧道衬砌裂缝检测算法［J］.铁道学报,2023,45(10):162-170. ZHOU Zhong,ZHANG Junjie,LIU Siping.Tunnel lining crack detection algorithm based on improved YOLOv4 ［J］.Journal of the China Railway Society,2023,45(10):162-170.
［7］王建锋,刘文豪,潘清云.基于图像处理与深度学习的隧道衬砌裂缝检测［J］.北京交通大学学报,2022,46(5):19-29. WANG Jianfeng,LIU Wenhao,PAN Qingyun.Crack detection for tunnel lining based on image processing and deep learning ［J］.Journal of Beijing Jiaotong University,2022,46(5):19-29.
［8］何兆益,常宝霞,吴逸飞,等.基于YOLO v5-IBX网络模型的公路隧道衬砌裂缝检测方法研究 ［J］.沈阳建筑大学学报(自然科学版),2023,39(5):888-898. HE Zhaoyi,CHANG Baoxia,WU Yifei,et al.Research on crack detection method of highway tunnel lining based on YOLO v5-lBX network model ［J］.Journal of Shenyang Jianzhu University (Natural Science),2023,39 (5):888-898.
［9］陈莹莹,刘新根,黄永亮,等.基于神经网络与边缘修正的隧道衬砌裂缝识别［J］.现代隧道技术,2022,59(6):24-34. CHEN Yingying,LIU Xin’gen,HUANG Yongliang,et al.Crack identification of tunnel lining based on neural network and edge correction ［J］.Modern Tunnelling Technology,2022,59(6):24-34.
［10］刘新根,陈莹莹,王雨萱,等.基于级联神经网络的隧道衬砌裂缝识别算法研究［J］.铁道学报,2021,43(10):127-135. LIU Xin’gen,CHEN Yingying,WANG Yuxuan,et al.Research on tunnel lining crack identification algorithm based on cascade neural network ［J］.Journal of the China Railway Society,2021,43(10):127-135.
［11］YU Z W,SHEN Y G,SHEN C K.Areal-time detection approach for bridge cracks based on YOLOv4-FPM［J］.Automation in Construction,2021,122:103514.
［12］ZHOU Z,SHANG J J,GONG C J.Automatic detection method of tunnel lining multi-defects via an enhanced you only look once network ［J］.Computer-Aided Civil and Infrastructure Engineering,2022,37(6):762-780.
［13］朱咏梅,李玉玲,奚峥皓，等.注意力可变形卷积网络的木质板材瑕疵识别［J］.西南大学学报(自然科学版),2024,46(2):159-169. ZHU Yongmei,LI Yuling,XI Zhenghao,et al.Attention deformable convolutional networks for wooden panel defect recognition ［J］.Journal of Southwest University(Natural Science Edition),2024,46(2):159-169.
［14］胡宏宇,张争光,曲优,等.基于双分支和可变形卷积网络的驾驶员行为识别方法［J/OL］.吉林大学学报(工学版),1-12［2024-02-25］.https://doi.org/10.13229/j.cnki.jdxbgxb.20230313.
［15］高涵,赵培培,于正,等.基于特征增强与Transformer的煤矿输送带异物检测［J/OL］.煤炭科学技术:1-11［2024-02-26］.http://kns.cnki.net/kcms/detail/11.2402.td.20240119.1515.012.html.
［16］XIONG Y W,LI Z Q,CHEN Y T,et al.Efficient deformable convnets:rethinking dynamic and sparse operator for vision applications ［J］.arXiv preprint,2024:2401.06197v1.

相似文献/References:

[1]安永林,蔡海滨,岳健,等.高铁对邻近矿区主井的动力安全影响分析[J].中国安全生产科学技术,2015,11(4):40.[doi:10.11731/j.issn.1673-193X.2015.04.006]
　AN Yong-lin,CAI Hai-bin,YUE Jian,et al.Analysis on influence of high-speed rail on main shaft of neighboring mine[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2015,11(8):40.[doi:10.11731/j.issn.1673-193X.2015.04.006]
[2]吴波,路明,雷领,等.基于正交试验、组合赋权-灰色关联度法研究隧道施工最优方案[J].中国安全生产科学技术,2019,15(8):124.[doi:10.11731/j.issn.1673-193x.2019.08.020]
　WU Bo,LU Ming,LEI ling,et al.Research on optimal scheme of tunnel construction based on orthogonal test and combined weightinggrey correlation method[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2019,15(8):124.[doi:10.11731/j.issn.1673-193x.2019.08.020]
[3]崔光耀,祁家所,王明胜.中义隧道片理化玄武岩段大变形控制技术研究*[J].中国安全生产科学技术,2020,16(10):115.[doi:10.11731/j.issn.1673-193x.2020.10.018]
　CUI Guangyao,QI Jiasuo,WANG Mingsheng.Study on large deformation control technology of schistositized basalt section in Zhongyi tunnel[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2020,16(8):115.[doi:10.11731/j.issn.1673-193x.2020.10.018]
[4]崔光耀,田宇航,麻建飞,等.高烈度地震区隧道软硬围岩交接段刚柔并济抗减震技术研究*[J].中国安全生产科学技术,2021,17(4):135.[doi:10.11731/j.issn.1673-193x.2021.04.022]
　CUI Guangyao,TIAN Yuhang,MA Jianfei,et al.Research on rigid-flexible combination anti-seismic technology on soft-hard surrounding rock intersection of tunnel in high intensity seismic area[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2021,17(8):135.[doi:10.11731/j.issn.1673-193x.2021.04.022]
[5]王兵.新建草莓沟2号隧道近接下穿既有隧道施工安全影响分析*[J].中国安全生产科学技术,2021,17(5):106.[doi:10.11731/j.issn.1673-193x.2021.05.016]
　WANG Bing.Analysis on construction safety influence of newly-built Caomeigou No.2 tunnel close undercrossing existing tunnel[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2021,17(8):106.[doi:10.11731/j.issn.1673-193x.2021.05.016]
[6]安栋,刘天旺,郭艳军.强震区跨断层隧道纤维混凝土衬砌抗震效果分析*[J].中国安全生产科学技术,2021,17(6):98.[doi:10.11731/j.issn.1673-193x.2021.06.016]
　AN Dong,LIU Tianwang,GUO Yanjun.Analysis on seismic effect of fiber concrete lining in cross-fault tunnel at strong earthquake area[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2021,17(8):98.[doi:10.11731/j.issn.1673-193x.2021.06.016]
[7]崔光耀,石文昊,王明胜,等.高烈度地震区跨断层隧道不同厚度减震层减震效果分析*[J].中国安全生产科学技术,2021,17(7):130.[doi:10.11731/j.issn.1673-193x.2021.07.021]
　CUI Guangyao,SHI Wenhao,WANG Mingsheng,et al.Analysis on shock absorption effect of shock-absorbing layer with different thicknesses for fault-crossing tunnels in high seismic intensity zone[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2021,17(8):130.[doi:10.11731/j.issn.1673-193x.2021.07.021]
[8]于阳,孙雅珍,林志军,等.基坑开挖引起邻侧盾构隧道开裂特性研究*[J].中国安全生产科学技术,2021,17(10):99.[doi:10.11731/j.issn.1673-193x.2021.10.015]
　YU Yang,SUN Yazhen,LIN Zhijun,et al.Study on cracking characteristics of adjacent shield tunnel caused by foundation pit excavation[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2021,17(8):99.[doi:10.11731/j.issn.1673-193x.2021.10.015]
[9]段在鹏,李帆,邱少辉,等.地铁沿线老旧房屋结构安全预警模型*[J].中国安全生产科学技术,2022,18(3):162.[doi:10.11731/j.issn.1673-193x.2022.03.025]
　DUAN Zaipeng,LI Fan,QIU Shaohui,et al.Early warning model for structural safety of old buildings along metro lines[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2022,18(8):162.[doi:10.11731/j.issn.1673-193x.2022.03.025]
[10]曹成威,石钰锋,詹涛,等.考虑动态施工超长管棚预支护力学特性及参数影响分析*[J].中国安全生产科学技术,2022,18(6):98.[doi:10.11731/j.issn.1673-193x.2022.06.015]
　CAO Chengwei,SHI Yufeng,ZHAN Tao,et al.Analysis on mechanical characteristics and parameter influence of pre-support for ultra-long pipe shed considering dynamic construction[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2022,18(8):98.[doi:10.11731/j.issn.1673-193x.2022.06.015]

备注/Memo

备注/Memo:
收稿日期： 2024-02-28；网络首发日期： 2024-08-01
* 基金项目： 陕西省交通运输厅交通科技项目（22-09K）
作者简介： 孙己龙，硕士，高级工程师，主要研究方向为公路工程质量和安全监督工作。
通信作者： 刘勇，博士研究生，主要研究方向为运营隧道病害智能检测及评价。
更新日期/Last Update: 2024-08-26