基于机器学习方法的路面病害检测研究综述

邹政; 陈江; 郎洪; 王笑风; 万晨光; 丁朔; 陆键

doi:10.3963/j.jssn.1674-4861.2025.02.016

基于机器学习方法的路面病害检测研究综述

doi: 10.3963/j.jssn.1674-4861.2025.02.016

邹政^1,,
陈江¹,
郎洪^1, ,,
王笑风²,
万晨光²,
丁朔¹,
陆键^{1, 3}

1.
同济大学道路与交通工程教育部重点实验室上海 201804
2.
河南省中工设计研究院集团股份有限公司郑州 450000
3.
山东交通学院济南 250357

基金项目:

国家自然科学基金项目 62206201

河南省交通运输厅科技项目 2023-1-1

中国博士后科学基金项目 2023M732644

详细信息

作者简介:
邹政(1993—)，博士研究生. 研究方向：智能交通系统工程. E-mail: zouzhn@tongji.edu.cn

通讯作者:
郎洪(1994—)，博士，副研究员. 研究方向：交通信息工程及控制、全域感知等. E-mail: honglang@to-gji.edu.cn

中图分类号: U416.2
计量
- 文章访问数: 43
- HTML全文浏览量: 20
- PDF下载量: 0
- 被引次数: 0
出版历程
- 收稿日期: 2024-05-24
- 网络出版日期: 2025-09-29

A Review of Pavement Distress Detection Based on Machine Learning Methods

1.
The Key Laboratory of Road and Traffic Engineering of Ministry of Education, Tongji University, Shanghai 201804, China
2.
Henan Zhonggong Design and Research Institute Group Co., Ltd., Zhengzhou 450000, China
3.
Shandong Jiaotong University, Jinan 250357, China

摘要

摘要: 道路路面的质量状况直接影响其使用寿命和行车安全。路面病害具有机理复杂、形态多样的特点，且在空间尺度上呈现出大样本分布，给度量与标注带来较大挑战，传统检测方法难以满足多场景、多类型病害识别的实际应用需求。本文回顾了机器学习方法在路面病害检测中的重要研究成果，比较了不同识别方法的技术原理和适用性，并总结了公开和私有的路面病害图像数据集。然后，基于三维深度数据对路面病害展开多种特征分析，为挖掘路面病害多元特征与人工智能模型的相互作用提供算法基础。最后，系统性介绍了通用的客观评测指标以保证路面病害智能算法评价的公正性。病害识别经历了传统的阈值分割、面元分类和目标检测并逐步发展到像素级图像分割的趋势，自动识别算法的准确性和通用性得到了大幅度提升。传统机器学习方法和深度学习方法之间仍然可以相互借鉴与融合，以覆盖更多的路面病害类型并促使模型能够学习到更多有效特征，从而提高病害检测精度及效率。现有不同识别算法应以运行时间、显存占比、运算量、识别率作为一致的性能评价指标，并考虑不同道路环境下不同维度的路面病害数据集。未来应研发适应复杂工况的检测与处理方法，提升信噪比与模型泛化能力，推动智能算法在多场景中的落地应用。
- 道路工程 /
- 病害检测 /
- 机器学习 /
- 像素分割 /
- 评价方法
Abstract: Road surface quality directly influences service life and driving safety. Pavement distress involves complex mechanisms, diverse forms, and large-scale spatial distribution, posing challenges for measurement and annotation. Traditional detection methods struggle to meet practical needs for identifying various distress types in diverse scenarios. This paper reviews achievements in machine learning-based pavement distress detection, compares principles and applicability of recognition methods, and summarizes public and private image datasets of pavement distress. Subsequent feature analyses of pavement distress using 3D depth data provide a basis for exploring interactions between multidimensional features and AI models. Finally, general objective evaluation metrics are systematically introduced to ensure fairness in AI-based pavement distress assessment. Distress recognition has evolved from threshold segmentation, element classification, and object detection to pixel-level segmentation, significantly improving accuracy and generalizability. Traditional machine learning and deep learning methods can be integrated to detect more distress types and learn more effective features, improving accuracy and efficiency. Current algorithms should be evaluated using consistent metrics, including runtime, memory usage, computational load, and recognition rate, while considering datasets with varying distress dimensions and road environments. Future work should develop detection and processing methods for complex conditions, enhance signal-to-noise ratios and generalization, and support deploying intelligent algorithms in diverse scenarios.
- road engineering /
- distress detection /
- machine learning /
- pixel segmentation /
- evaluation method

HTML全文

图 1 深度学习对路面病害图像的处理方式

Figure 1. The processing approach of pavement distress images using deep learning

下载: 全尺寸图片幻灯片

图 2 8个裂缝类路面病害数据集

Figure 2. Eight datasets of crack-type pavement distresses

下载: 全尺寸图片幻灯片

图 3 3个多类型路面病害数据集

Figure 3. Three datasets of multiple types of pavement distresses

下载: 全尺寸图片幻灯片

图 4 8组不同的路面病害三维图像

Figure 4. Eight sets of different 3D images of pavement distresses

下载: 全尺寸图片幻灯片

图 5 裂缝类路面病害的横断面高程曲线

Figure 5. Transverse elevation profiles of crack-type pavement distresses

下载: 全尺寸图片幻灯片

图 6 变形类路面病害的横断面高程曲线

Figure 6. Transverse elevation profiles of deformation-type pavement distresses

下载: 全尺寸图片幻灯片

图 7 路面病害三维图像的灰度直方图（图 4数据）

Figure 7. Grayscale histogram of 3D pavement distress images (data from Fig. 4)

下载: 全尺寸图片幻灯片

图 8 路面病害三维图像的灰度等高线（图 4数据）

Figure 8. Grayscale contour lines of 3D pavement distress images (data from Fig. 4)

下载: 全尺寸图片幻灯片

图 9 路面病害三维图像的傅立叶频域分析结果（图 4数据）

Figure 9. Fourier frequency domain analysis results of 3D pavement distress images(data from Fig. 4)

下载: 全尺寸图片幻灯片

图 10 Canny边缘检测算子检测结果（图 4数据）

Figure 10. Detection results using the Canny edge detection operator (data from Fig. 4)

下载: 全尺寸图片幻灯片

表 1 公开的主流路面病害数据集的基本信息

Table 1. Basic information of publicly available mainstream pavement distress datasets

类型	数据集	数量	尺寸	光源	相机	标注方式
裂缝类病害	CrackTree260	260	800×600	可见光	面阵	像素级
	Crack500	500	2 000×1 500	可见光	手机	像素级
	CrackLS315	315	512×512	激光	线阵	像素级
	CRKWH100	100	512×512	可见光	线阵	像素级
	CFD	118	480×320	可见光	手机	像素级
	AigleRN	38	991×462 311×462	可见光	面阵	像素级
	GAPs384	384	1 920×1 080	激光	线阵	像素级
	AIMCrack	527	1 920×1 080	可见光	行车记录仪	像素级
多种病害类型	GAPs	1 969	1 920×1 080	激光	线阵	边界框
	GAPs v2	2 468	1 920×1 080	激光	线阵	边界框
	Road Damage	9 053	600×600	可见光	手机	边界框
	ISTD-PDS7	18527		HID电灯	面阵	像素级/二值

下载: 导出CSV

表 2 三维图像全局纹理特征

Table 2. Global texture features of 3D images

图像序号	类型	相对深度	平均对比度	平滑度	偏态性	一致性	熵值	$\left\\|\boldsymbol{T}_i-\overline{\boldsymbol{T}}\right\\|$
A1	完好路面	0.212 6	0.107	0.381	0.680 4	0.601 4	0.248 2	0.182 7
A2	完好路面	0.493 9	0.364 9	0.761 9	0.619 5	0.210 1	0.682	0.328 5
B1	松散类	1	0.306 1	0.714 3	0	0.956 5	0.067 5	1
B2	松散类	0.658 3	0.323	0.714 3	0.713	0.239 1	0.627 7	0.052 9
C1	线裂	0.352	0	0	0.746 7	0.623 2	0.218 7	0.113 4
C2	线裂	0.091 6	0.654 2	0.904 8	0.850 8	0.058	0.849 4	0.371 8
D1	网状裂缝	0.1	0.460 7	0.809 5	1	0.268 1	0.653 2	0.586 5
D2	网状裂缝	0	0.334 7	0.714 3	0.793 3	0.340 6	0.497 9	0.254
E1	变形类	0.405 4	0.720 3	0.952 4	0.491 5	0	1	0.530 7
E2	变形类	0.391 1	1	1	0.196 3	0.152 2	0.850 9	0.841 5
F1	沥青修补	0.555 3	0.162 5	0.476 2	0.678 5	0.485 5	0.353 8	0.189
F2	沥青修补	0.432 8	0.141 2	0.428 6	0.857 9	0.644 9	0.317 9	0.384 4
G1	白色标线	0.527 2	0.541 2	0.857 1	0.569 1	0.188 4	0.722 8	0.418 1
G2	白色标线	0.333 1	0.362 6	0.761 9	0.727	0.195 7	0.708 1	0
H1	桥接缝	0.515 2	0.236 4	0.619	0.552 7	0.405 8	0.479 2	0.443 8
H2	桥接缝	0.913 6	0.208 3	0.571 4	0.299 2	1	0	0.746 6

下载: 导出CSV

参考文献(54)

[1]	HUANG J, LIU W, SUN X. A pavement crack detection method combining 2D with 3D information based on Dempster-Shafer theory[J]. Computer-aided Civil and Infrastructure Engineering, 2014, 29(4): 299-313.
[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]	马建, 赵祥模, 贺拴海, 等. 路面检测技术综述[J]. 交通运输工程学报, 2017, 17(5): 121-137. MA J, ZHAO X M, HE S H, et al. Review of pavement detection technology[J]. Journal of Traffic and Transportation Engineering, 2017, 17(5): 121-137(. in Chinese)
[4]	陈江, 原野, 郎洪, 等. 基于多分支深度学习的沥青路面多病害检测方法[J]. 东南大学学报(自然科学版), 2023, 53 (1): 123-129. CHEN J, YUAN Y, LANG H, et al. Multi-distress detection method for asphalt pavements based on multi-branch deep learning[J]. Journal of Southeast University(Natural Science Edition), 2023, 53(1): 123-129(. in Chinese)
[5]	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.
[6]	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.
[7]	LANG H, LU J J, LOU Y, et al. Pavement cracking detection and classification based on 3D image using multiscale clustering model[J]. Journal of Computing in Civil Engineering, 2020, 34(5): 04020034.
[8]	王艾迪, 彭一川, 郎洪, 等. 基于YOLOX-Transformer两步模型的路面坑槽提取方法[J]. 中国公路学报, 2023, 36(12): 304-317. WANG A D, PENG Y C, LANG H, et al. Pavement pothole extraction based on YOLOX-Transformer two-step model[J]. China Journal of Highway and Transport, 2023, 36(12): 304-317(. in Chinese)
[9]	青晨, 禹晶, 肖创柏, 等. 深度卷积神经网络图像语义分割研究进展[J]. 中国图象图形学报, 2020, 25(6): 1069-1090. QING C, YU J, XIAO C B, et al. Deep convolutional neural network for semantic image segmentation[J]. Journal of Image and Graphics, 2020, 25(6): 1069-1090(. in Chinese)
[10]	GAVILÁN M, BALCONES D, MARCOS O, et al. Adaptive road crack detection system by pavement classification[J]. Sensors, 2011, 11(10): 9628-9657.
[11]	SHI Y, CUI L, QI Z, et al. Automatic road crack detection using random structured forests[J]. IEEE Transactions on Intelligent Transportation Systems, 2016, 17(12): 3434-3445.
[12]	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. LANG H, WEN T, LU J, et al. 3D pavement crack detection method based on deep learning[J]. Journal of Southeast University(Natural Science Edition), 2021, 51(1): 53-60. (in Chinese)
[13]	郭立媛, 张磊, 李威, 等. 基于先验知识MinMax k-Means聚类算法的道路裂缝研究[J]. 中国测试, 2018, 44(4): 112-117. GUO L Y, ZHANG L, LI W, et al. Research on road crack based on MinMax k-Means clustering algorithm with prior knowledge[J]. China Measurement & Test, 2018, 44(4): 112-117(. in Chinese)
[14]	ABDELMAWLA A, YANG J J, KIM S S. Unsupervised learning of pavement distresses from surface images[C]. Civil Infrastructures Confronting Severe Weathers and Climate Changes Conference, Cham, Switzerland: Springer International Publishing, 2021.
[15]	GUO N, YOU L, LONG Z, et al. Computationally-affordable unsupervised machine learning algorithm to identify the level of distress severity in pavement functional performance[J]. IEEE Transactions on Intelligent Transportation Systems, 2023, 24(7): 7342-7356.
[16]	OLIVEIRA H, CORREIA P L. Automatic road crack detection and characterization[J]. IEEE Transactions on Intelligent Transportation Systems, 2012, 14(1): 155-168.
[17]	KHERADMANDI N, MEHRANFAR V. A critical review and comparative study on image segmentation-based techniques for pavement crack detection[J]. Construction and Building Materials, 2022, 321: 126162.
[18]	左永霞. 高速公路路面破损图像识别技术研究[D]. 长春: 吉林大学, 2008. ZUO Y X. Research on highway pavement surface distress image recognition[D]. Changchun: Jilin University, 2008(. in Chinese)
[19]	HOANG N D, NGUYEN Q L. A novel method for asphalt pavement crack classification based on image processing and machine learning[J]. Engineering with Computers, 2019, 35: 487-498.
[20]	VYAS V, SINGH A P, SRIVASTAVA A. Prediction of asphalt pavement condition using FWD deflection basin parameters and artificial neural networks[J]. Road Materials and Pavement Design, 2021, 22(12): 2748-2766.
[21]	AHMADI A, KHALESI S, GOLROO A. An integrated machine learning model for automatic road crack detection and classification in urban areas[J]. International Journal of Pavement Engineering, 2022, 23(10): 3536-3552.
[22]	HSIEH Y A, TSAI Y J. Machine learning for crack detection: review and model performance comparison[J]. Journal of Computing in Civil Engineering, 2020, 34(5): 04020038.
[23]	WEN T, LANG H, DING S, et al. PCDNet: seed operation-based deep learning model for pavement crack detection on 3D asphalt surface[J]. Journal of Transportation Engineering, Part B: Pavements, 2022, 148(2): 04022023.
[24]	ZHANG K, CHENG H D, ZHANG B. Unified approach to pavement crack and sealed crack detection using preclassification based on transfer learning[J]. Journal of Computing in Civil Engineering, 2018, 32(2): 04018001.
[25]	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 A M, TONG Z, GAO J. Recognition and measurement of pavement disasters based on convolutional neural networks[J]. China Journal of Highway and Transport, 2018, 31 (1): 1-10(. in Chinese)
[26]	LEI X, LIU C, LI L, et al. Automated pavement distress detection and deterioration analysis using street view map[J]. IEEE Access, 2020, (8): 76163-76172.
[27]	DU Y, PAN N, XU Z, et al. Pavement distress detection and classification based on YOLO network[J]. International Journal of Pavement Engineering, 2021, 22(13): 1659-1672.
[28]	MAJIDIFARD H, JIN P, ADU-GYAMFI Y, et al. Pavement image datasets: a new benchmark dataset to classify and densify pavement distresses[J]. Transportation Research Record, 2020, 2674(2): 328-339.
[29]	YAO G, SUN Y, WONG M, et al. A real-time detection method for concrete surface cracks based on improved YOLOv4[J]. Symmetry, 2021, 13(9): 1716.
[30]	IBRAGIMOV E, LEE H J, LEE J J, et al. Automated pavement distress detection using region based convolutional neural networks[J]. International Journal of Pavement Engineering, 2022, 23(6): 1981-1992.
[31]	ZHU J, ZHONG J, MA T, et al. Pavement distress detection using convolutional neural networks with images captured via UAV[J]. Automation in Construction, 2022, 133: 103991.
[32]	GUO K, HE C, YANG M, et al. A pavement distresses identification method optimized for YOLOv5s[J]. Scientific Reports, 2022, 12(1): 3542.
[33]	YANG X, LI H, YU Y, et al. Automatic pixel-level crack detection and measurement using fully convolutional network[J]. Computer-aided Civil and Infrastructure Engineering, 2018, 33(12): 1090-1109.
[34]	张涛, 王金, 刘斌, 等. 基于改进U-net的沥青路面图像裂缝分割方法[J]. 交通信息与安全, 2023, 41(6): 90-99. ZHANG T, WANG J, LIU B, et al. Crack segmentation of asphalt pavement images based on improved U-net[J]. Journal of Transport Information and Safety, 2023, 41(6): 90-99(. in Chinese)
[35]	BANG S, PARK S, KIM H, et al. Encoder-decoder network for pixel-level road crack detection in black-box images[J]. Computer-aided Civil and Infrastructure Engineering, 2019, 34(8): 713-727.
[36]	HUYAN J, LI W, TIGHE S, et al. CrackU-net: a novel deep convolutional neural network for pixelwise pavement crack detection[J]. Structural Control and Health Monitoring, 2020, 27(8): 2551.
[37]	WEN T, DING S, LANG H, et al. Automated pavement distress segmentation on asphalt surfaces using a deep learning network[J]. International Journal of Pavement Engineering, 2023, 24(2): 2027414.
[38]	LIU J, YANG X, LAU S, et al. Automated pavement crack detection and segmentation based on two-step convolutional neural network[J]. Computer-aided Civil and Infrastructure Engineering, 2020, 35(11): 1291-1305.
[39]	FEI Y, WANG K C P, ZHANG A, et al. Pixel-level cracking detection on 3D asphalt pavement images through deep-learning-based CrackNet-V[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 21(1): 273-284.
[40]	TAN C, UDDIN N, MOHAMMED Y M. Deep learning-based crack detection using mask R-CNN technique[C]. 9th International Conference on Structural Health Monitoring of Intelligent Infrastructure, Saint Louis, USA: NSF Public Access Repository, 2019.
[41]	ZHANG A, WANG K C P, FEI Y, et al. Automated pixel-level pavement crack detection on 3D asphalt surfaces with a recurrent neural network[J]. Computer-aided Civil and Infrastructure Engineering, 2019, 34(3): 213-229.
[42]	SHIM S. Self-training approach for crack detection using synthesized crack images based on conditional generative adversarial network[J]. Computer-aided Civil and Infrastructure Engineering, 2024, 39(7): 1019-1041.
[43]	ZOU Q, CAO Y, LI Q, et al. Cracktree: automatic crack detection from pavement images[J]. Pattern Recognition Letters, 2012, 33(3): 227-238.
[44]	YANG F, ZHANG L, YU S, et al. Feature pyramid and hierarchical boosting network for pavement crack detection[J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 21(4): 1525-1535.
[45]	ZOU Q, ZHANG Z, LI Q, et al. Deepcrack: learning hierarchical convolutional features for crack detection[J]. IEEE Transactions on Image Processing, 2018, 28(3): 1498-1512.
[46]	CHAMBON S, MOLIARD J M. Automatic road pavement assessment with image processing: review and comparison[J]. International Journal of Geophysics, 2011, 2011(1): 989354.
[47]	STRICKER R, EISENBACH M, SESSELMANN M, et al. Improving visual road condition assessment by extensive experiments on the extended gaps dataset[C]. 2019 International Joint Conference on Neural Networks(IJCNN), Budapest, Hungary: IEEE, 2019.
[48]	MAEDA H, SEKIMOTO Y, SETO T, et al. Road damage detection and classification using deep neural networks with smartphone images[J]. Computer-aided Civil and Infrastructure Engineering, 2018, 33(12): 1127-1141.
[49]	SONG W, ZHANG Z, ZHANG B, et al. ISTD-PDS7: a benchmark dataset for multi-type pavement distress segmentation from CCD images in complex scenarios[J]. Remote Sensing, 2023, 15(7): 1750.
[50]	SOME L. Automatic image-based road crack detection methods[D]. Stockholm: KTH Royal Institute of Technology, 2016.
[51]	ZHANG L, YANG F, ZHANG Y D, et al. Road crack detection using deep convolutional neural network[C]. 2016 IEEE International Conference on Image Processing(ICIP), Phoenix, USA: IEEE, 2016.
[52]	ZHANG A, WANG K C P, LI B, 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.
[53]	GONG H, LIU L, LIANG H, et al. A state-of-the-art survey of deep learning models for automated pavement crack segmentation[J]. International Journal of Transportation Science and Technology, 2024, 13: 44-57.
[54]	徐志刚. 基于多特征融合的路面破损图像自动识别技术研究[D]. 西安: 长安大学, 2012. XU Z G. Study on the automatic identification technology for pavement distress image based on multi-features fusion[D]. Xi'an: Chang'an University, 2012(. in Chinese