沥青路面内部裂缝雷达图像智能判识算法研究

doi:10.6040/j.issn.1672-3961.0.2024.092

摘要/Abstract

摘要： 采用探地雷达对典型路段的路面裂缝识别并定位。通过钻芯取样与铣刨观察结合的方法验证,构建了包含728张雷达图像的数据库;采用YOLO v8l算法学习裂缝特征,通过在YOLO v8l算法的基础上引入注意力机制和修改激活函数,克服了路面裂缝图像特征多变、噪声杂波明显等对于智能判识造成的干扰,同时消除了模型的过拟合现象;对算法修正后,模型的计算参数增多,计算效率提升,修正算法的识别精确度和召回率分别达到99.4%和92.3%。训练过程中平均精度均值与损失函数的震荡幅度较小,表示该数据集标注原则统一,证明了采用该方法识别路面裂缝的有效性与可靠性。

关键词: 探地雷达, 人工智能, 目标检测, 深度学习算法, 卷积神经网络

Abstract: This study employed GPR(ground penetrating radar)to identify and locate surface cracks in typical road segments. The method combined core drilling sampling and milling observation for validation, resulting in the construction of a database containing 728 radar images. The YOLO v8l algorithm was used to learn crack features. By incorporating an attention mechanism and modifying the activation function within the YOLO v8l framework, the study overcame the interference caused by the variability of road crack image features and significant noise, while also eliminating model overfitting. After modifying the algorithm, the model's computational parameters increased, and the computational efficiency improved. The precision and recall rates of the revised algorithm reached 99.4% and 92.3%, respectively. During training, the mean average precision and loss function fluctuations were minimal, indicating that the dataset annotation principles were consistent. This proved the effectiveness and reliability of the proposed method for identifying road surface cracks.

Key words: ground penetrating radar, artificial intelligence, object detection, deep learning algorithm, convolutional neural network

中图分类号:

U418.6

董明书,陈俐企,马川义,张珠皓,孙仁娟,管延华,庄培芝. 沥青路面内部裂缝雷达图像智能判识算法研究[J]. 山东大学学报 (工学版), 2025, 55(3): 72-79.

DONG Mingshu, CHEN Liqi, MA Chuanyi, ZHANG Zhuhao, SUN Renjuan, GUAN Yanhua, ZHUANG Peizhi. Deep learning-based intelligent judgment for radar detection of pavement cracks[J]. Journal of Shandong University(Engineering Science), 2025, 55(3): 72-79.

参考文献

[1] 黄哲骁, 陶毅荣. 基于三维探地雷达的襄荆高速路面病害现状分析[J]. 交通节能与环保, 2024, 20(2): 151-156. HUANG Zhexiao, TAO Yirong. Analysis of pavement disease status of Xiangjing expressway based on three-dimensional ground-penetrating radar[J]. Transport Energy Conservation & Environmental Protection, 2024, 20(2): 151-156.
[2] 周陈婴. 隧道衬砌检测探地雷达图像分析与工程应用[J]. 山东大学学报(工学版), 2018, 48(4): 61-68. ZHOU Chenying. Image analysis and engineering application of ground penetrating radar in tunnel lining detection[J]. Journal of Shandong University(Engi-neering Science), 2018, 48(4): 61-68.
[3] 周轮, 李术才, 许振浩, 等. 隧道综合超前地质预报技术及其工程应用[J]. 山东大学学报(工学版), 2017, 47(2): 55-62. ZHOU Lun, LI Shucai, XU Zhenhao, et al. Integrated advanced geological prediction technology of tunnel and its engineering application[J]. Journal of Shandong University(Engineering Science), 2017, 47(2): 55-62.
[4] 高阳, 张庆松, 原小帅, 等. 地质雷达在岩溶隧道超前预报中的应用[J]. 山东大学学报(工学版), 2009, 39(4): 82-86. GAO Yang, ZHANG Qingsong, YUAN Xiaoshuai, et al. Application of geological radar to geological forecast in Karst tunnel[J]. Journal of Shandong University(Engineering Science), 2009, 39(4): 82-86.
[5] 罗诗光, 任强, 王成浩, 等. 基于深度学习的GPR时频域联合电磁反演方法[J]. 电波科学学报, 2022, 37(4): 555-567. LUO Shiguang, REN Qiang, WANG Chenghao, et al. GPR time-frequency domain joint electromagnetic inversion method based on deep learning[J]. Chinese Journal of Radio Science, 2022, 37(4): 555-567.
[6] 赵勇. 基于CNN的探地雷达数据反演与道路病害自动识别[D]. 长春: 吉林大学, 2022. ZHAO Yong. Research on inversion of ground penetrating radar data and automatic detection of road diseases based on CNN [D]. Changchun: Jilin University, 2022.
[7] 郑金杏. 探地雷达层状介质参数反演算法研究[D]. 桂林:桂林电子科技大学, 2023. ZHENG Jinxing. Research on the inversion algorithm for layered medium parameters in ground penetrating radar[D]. Guilin: Guilin University of Electronic Techno-logy, 2023.
[8] CARION N, MASSA F, SYNNAEVE G, et al. End-to-end object detection with transformers[C] // European Conference on Computer Vision. Cham, Switzerland: Springer International Publishing, 2020: 213-229.
[9] QIN Z Y, LU X K, NIE X S, et al. Exposing the self-supervised space-time correspondence learning via graph kernels[C] // Proceedings of the AAAI Conference on Artificial Intelligence. Washington, D.C., USA: AAAI Press, 2023: 2110-2118.
[10] ZHANG X M, ZHAO P, JI J S, et al. Video corpus moment retrievalvia deformable multigranularity feature fusion and adversarial training[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2024, 34(8): 6686-6698.
[11] DINH K, GUCUNSKI N, DUONG T H. An algorithm for automatic localization and detection of rebars from GPR data of concrete bridge decks[J]. Automation in Construction, 2018, 89: 292-298.
[12] OZKAYA U, MELGANI F, BELETE BEJIGA M, et al. GPR B-scan image analysis with deep learning methods[J]. Measurement, 2020, 165: 107770.
[13] ME?瘙塁ECAN(·overI), BETIM Ç,(·overI)HSAN Ö B. Feature vector for underground object detection using B-scan images from GprMax[J]. Microprocessors and Microsystems, 2020, 76: 103116.
[14] 薛丽君. 基于探地雷达B-scan图像双曲线特征的目标检测与定位研究[D]. 哈尔滨: 哈尔滨工业大学, 2021. XUE Lijun. Research on target detection and location based on hyperbolic-shaped signatures in GPR B-Scan image [D]. Harbin: Harbin Institute of Techno-logy, 2021.
[15] 童峥. 基于深度学习和探地雷达技术的路面结构病害检测研究[D]. 西安: 长安大学, 2018. TONG Zheng. Research on pavement distress inspection based on deep learning and ground penetrating radar [D]. Xi'an: Changan University, 2018.
[16] 胡浩帮, 方宏远, 王复明, 等. 基于Faster R-CNN算法的探地雷达管线目标智能识别[J]. 城市勘测, 2020(3): 203-208. HU Haobang, FANG Hongyuan, WANG Fuming, et al. Intelligent recognition of pipeline target based on Faster R-CNN algorithm for ground penetrating radar[J]. Urban Geotechnical Investigation & Surveying, 2020(3): 203-208.
[17] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C] //2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Las Vegas, USA: IEEE, 2016: 779-788.
[18] 覃紫馨, 姜彦南, 徐立, 等. 基于YOLO算法的探地雷达道路图像异常自动检测[J]. 科学技术与工程, 2023, 23(27): 11505-11512. QIN Zixin, JIANG Yannan, XU Li, et al. Automatic detection of anomalies in GPR images based on YOLO algorithm[J]. Science Technology and Engineering, 2023, 23(27): 11505-11512.
[19] 杨洋, 赵广茂, 张志厚, 等. 铁路路基病害探地雷达智能识别方法[J]. 地球物理学进展, 2024, 39(6): 2471-2482. YANG Yang, ZHAO Guangmao, ZHANG Zhihou, et al. Intelligent detection method for railway subgrade diseases based on ground-penetrating radar[J]. Progress in Geophysics, 2024, 39(6): 2471-2482.
[20] 赵镇, 黄勇, 冯昆. 三维探地雷达在道路检测中的研究与应用[J]. 测绘通报, 2024(2): 148-152. ZHAO Zhen, HUANG Yong, FENG Kun. Research and application of 3D ground penetrating in road detection[J]. Bulletin of Surveying and Mapping, 2024(2): 148-152.
[21] 范剑伟. 基于无损检测技术的沥青路面裂缝及结构状态检测评价方法研究[D]. 南京: 东南大学, 2022. FAN Jianwei. Research on detection and evaluation method of asphalt pavement cracks and structural state based on nondestructive testing technology[D]. Nanjing: Southeast University, 2022.
[22] BARKATAKI N, TIRU B, SARMA U. A CNN model for predicting size of buried objects from GPR B-Scans[J]. Journal of Applied Geophysics, 2022, 200: 104620.
[23] 周黎明, 张杨, 付代光, 等. 道路地下空洞探地雷达波场和时频特性[J]. 同济大学学报(自然科学版), 2024, 52(1): 77-85. ZHOU Liming, ZHANG Yang, FU Daiguang, et al. Wave field and time-frequency characteristics of ground penetrating radar for underground cavity of road[J]. Journal of Tongji University(Natural Science), 2024, 52(1): 77-85.
[24] LU H P, PLATANIOTIS K N, VENETSANOPOULOS A N. MPCA: multilinear principal component analysis of tensor objects[J]. IEEE Transactions on Neural Net-works, 2008, 19(1): 18-39.
[25] ZHU L, WANG X J, KE Z H, et al. BiFormer: vision transformer with bi-level routing attention[C] //2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). Vancouver, Canada: IEEE, 2023: 10323-10333.
[26] 常致富, 周风余, 王玉刚, 等. 基于深度学习的图像自动标注方法综述[J]. 山东大学学报(工学版), 2019, 49(6): 25-35. CHANG Zhifu, ZHOU Fengyu, WANG Yugang, et al. A survey of image captioning methods based on deep learning[J]. Journal of Shandong University(Engi-neering Science), 2019, 49(6): 25-35.
[27] KLAMBAUER G, UNTERTHINER T, MAYR A, et al. Self-normalizing neural networks[C] // Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach, USA: [s.n.] , 2017: 971-980.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed