FCM分割下的高架桥梁铺面病害识别
Identification of Pavement Damages on Elevated Bridges Using FCM Segmentation
为应对高架桥梁铺面病害存在的重叠与断裂问题,实现病害精准归类,本文提出一种高架桥梁铺面病害识别方法。该方法利用无人机巡检技术获取完整、连续的铺面厘米级病害域数据,经FCM分割并按相似属性分组后,结合快速区域卷积神经网络算法将病害域划分为不同区域,每个区域对应特定病害类别或正常铺面,从而实现病害识别。本研究对城市主干线高架桥铺面开展了识别实验验证,结果表明:该方法可有效克服复杂背景干扰,精准分割病害区域;对坑槽及细微裂缝等病害纹理具有较高的识别准确度与置信度;检测结果与病害实际区域吻合良好,具备优异的病害识别能力。
To achieve the classification of pavement defects on elevated bridges, particularly addressing the issues of overlapping and fracture, an identification method is proposed. Unmanned Aerial Vehicle (UAV) inspection technology is employed to acquire complete and continuous centimeter-level data of the pavement defect domain. Following Fuzzy C-Means (FCM) segmentation, the data are grouped by similar attributes and partitioned into distinct regions using a Fast Region-based Convolutional Neural Network (Fast R-CNN) algorithm; each group corresponds to a specific defect category or normal pavement condition for identification. Experiments conducted on the pavement of urban trunk line elevated bridges demonstrate that the proposed method effectively handles complex backgrounds, accurately segments various defect regions, and precisely identifies texture alignment in real potholes and micro-cracks. The approach exhibits high detection accuracy and confidence, showing strong consistency with actual defect areas and confirming its capability for accurate defect identification.
| [1] |
冯东明,余星宇,黎剑安,基于无人机的悬索桥主缆自动巡检及小样本表观病害识别[J].中国公路学报,2024,37(2):29-39. |
| [2] |
FENG Dongming,YU Xingyu,LI Jian'an,et al.Automatic inspection of main cables of suspension bridge based on UAV and apparent defect identification with small-sized samples[J].China Journal of Highway and Transport,2024,37(2): 29-39.(in Chinese) |
| [3] |
DJENOURI Y,BELHADI A.HOUSSEIN E H,et al.Intelligent graph convolutional neural network for road crack detection[J].IEEE Transactions on Intelligent Transportation Systems,2023,24(8):8475-8482. |
| [4] |
ALAMDARI A G,EBRAHIMKHANLOU A.A multi-scale robotic approach for precise crack measurement in concrete structures[J].Automation in Construction,2024,158(2):105215.1-105215.12. |
| [5] |
王桂平,陈旺桥,杨建喜,基于迁移学习的桥梁表观病害检测技术研究[J].铁道科学与工程学报,2022,19(6):1638-1646. |
| [6] |
WANG Guiping,CHEN Wangqiao,YANG Jianxi,et al.Research on bridge surface disease detection technology based on transfer learning[J].Journal of Railway Science and Engineering,2022,19(6):1638-1646.(in Chinese) |
| [7] |
彭雨诺,刘敏,万智,基于改进YOLO的双网络桥梁表观病害快速检测算法[J].自动化学报,2022,48(4):1018-1032. |
| [8] |
PENG Yunuo,LIU Min,WAN Zhi,et al.A fast detection algorithm for apparent diseases of dual network bridges based on improved YOLO[J].Journal of Automation,2022,48(4):1018-1032.(in Chinese) |
| [9] |
鲜荣,刘天成,朱超,基于YOLOv5和DeepLabv3+的桥梁结构病害智能识别技术[J].公路,2023,68(12):368-374. |
| [10] |
XIAN Rong,LIU Tiancheng,ZHU Chao,et al.Intelligent recognition technology for bridge structural diseases based on YOLOv5 and DeepLabv3+[J].Highway,2023,68(12):368-374.(in Chinese) |
| [11] |
余加勇,李锋,薛现凯,基于无人机及Mask R-CNN的桥梁结构裂缝智能识别[J].中国公路学报,2021,34(12):80-90. |
| [12] |
YU Jiayong,LI Feng,XUE Xiankai,et al.Intelligent identification of cracks in bridge structures based on drones and Mask R-CNN[J].Chinese Journal of Highways,2021,34(12):80-90.(in Chinese) |
| [13] |
袁依文,雷斌.无人机视觉技术在道路裂缝分类中的应用[J].机械设计与制造,2022(5):235-239. |
| [14] |
YUAN Yiwen,LEI Bin.Application of UAV vision technology in road crack classification[J].Mechanical Design and Manufacturing,2022(5):235-239.(in Chinese) |
| [15] |
张瑞,袁博,杨明,基于深度学习LoFTR算法的路面图像拼接[J].计算机系统应用,2023,32(2):170-180. |
| [16] |
ZHANG Rui,YUAN Bo,YANG Ming,et al.Road image stitching based on deep learning LoFTR algorithm[J].Computer System Applications,2023,32(2):170-180.(in Chinese) |
| [17] |
王耀东,朱力强,余祖俊,基于样本自动标注的隧道裂缝病害智能识别[J].西南交通大学学报,2023,58(5):1001-1008,1036. |
| [18] |
WANG Yaodong,ZHU Liqiang,YU Zujun,et al.Intelligent recognition of tunnel crack diseases based on sample automatic annotation[J].Journal of Southwest Jiaotong University,2023,58(5):1001-1008,1036.(in Chinese) |
| [19] |
薛冕,刘翔,石蕴玉,抑制光照不均匀影响的图像特征点提取算法[J].计算机工程与应用,2021,57(4):183-191. |
| [20] |
XUE Mian,LIU Xiang,SHI Yunyu,et al.Image feature point extraction algorithm for suppressing the influence of uneven lighting[J].Computer Engineering and Applications,2021,57(4):183-191.(in Chinese) |
| [21] |
徐少平,林珍玉,张贵珍,采用深度学习与图像融合混合实现策略的低照度图像增强算法[J].电子学报,2021,49(1):72-76. |
| [22] |
XU Shaoping,LIN Zhenyu,ZHANG Guizhen,et al.Low light image enhancement algorithm using a hybrid implementation strategy of deep learning and image fusion[J].Acta Sinica,2021,49(1):72-76.(in Chinese) |
| [23] |
王雷,刘小芳,赵良军,基于改进FCM算法的遥感影像滑坡变化检测方法[J].计算机仿真,2021,38(10):435-441. |
| [24] |
WANG Lei,LIU Xiaofang,ZHAO Liangjun,et al.Remote sensing image landslide change detection method based on improved FCM algorithm[J].Computer Simulation,2021,38(10):435-441.(in Chinese) |
| [25] |
晏班夫,徐观亚,栾健,基于Faster R-CNN与形态法的路面病害识别[J].中国公路学报,2021,34(9):181-193. |
| [26] |
YAN Banfu,XU Guanya,LUAN Jian,et al.Road surface disease recognition based on faster R-CNN and morphology method[J].Chinese Journal of Highways,2021,34(9):181-193.(in Chinese) |
/
| 〈 |
|
〉 |