采用多尺度特征增强的路面病害检测模型

胡鹏; 夏晓华; 钟预全; 段智威; 姚运仕; 成高立

doi:10.7652/xjtuxb202502016

您当前的位置：

首页 >

文章列表页 >

采用多尺度特征增强的路面病害检测模型

更新时间：2026-01-06

- 采用多尺度特征增强的路面病害检测模型
- Pavement Disease Detection Model Based on Multi-Scale Feature Reinforcement
- 西安交通大学学报 2025年59卷第2期页码：156-169
- 作者机构：
  
  1.长安大学道路施工技术与装备教育部重点实验室， 710064，西安
  2.陕西高速机械化工程有限公司， 710038，西安
- 作者简介：
  
  胡鹏（1999—），男，硕士生；
  夏晓华（通信作者），男，副教授，博士生导师。
- 基金信息：
  
  陕西省重点研发计划资助项目(2024GX-YBXM-197);陕西省交通运输厅交通科研项目(24-74K);秦创原引用高层次创新创业人才项目(QCYRCXM-2022-352)
- DOI：10.7652/xjtuxb202502016
  中图分类号： TP277
- 收稿：2024-06-21，
  
  网络首发：2024-08-30，
  
  纸质出版：2025-02-10
- 稿件说明：
移动端阅览
胡鹏, 夏晓华, 钟预全, 等. 采用多尺度特征增强的路面病害检测模型[J]. 西安交通大学学报, 2025,59(2):156-169.

HU Peng, XIA Xiaohua, ZHONG Yuquan, et al. Pavement Disease Detection Model Based on Multi-Scale Feature Reinforcement[J]. Journal of Xi’an Jiaotong University, 2025, 59(2): 156-169.
胡鹏, 夏晓华, 钟预全, 等. 采用多尺度特征增强的路面病害检测模型[J]. 西安交通大学学报, 2025,59(2):156-169. DOI： 10.7652/xjtuxb202502016.

HU Peng, XIA Xiaohua, ZHONG Yuquan, et al. Pavement Disease Detection Model Based on Multi-Scale Feature Reinforcement[J]. Journal of Xi’an Jiaotong University, 2025, 59(2): 156-169. DOI： 10.7652/xjtuxb202502016.

摘要

针对现有网络多尺度特征提取能力不足造成路面病害因尺寸差异难以完全识别的问题，提出了一种多尺度特征增强的路面病害检测模型。构建基于混合空洞卷积的快速空间金字塔池化模块，通过堆叠不同膨胀系数的空洞卷积进一步扩大网络感受野，以实现更大范围上下文信息的捕捉，并保留更多的空间信息；设计多路径特征融合网络，通过多分支和跳跃连接实现跨层级的特征捕捉，并减少特征融合过程中的信息丢失；采用

-means聚类算法结合交叉比获得合理的瞄点框；在损失函数中，设计一种面积惩罚项并设置下降梯度，提高预测框回归精度与效率；通过引入跨通道交互的高效注意力实现模型重要通道间的交互。实验结果表明：所提模型的检测精度比原模型YOLOv5s提高了4.0%；与Faster R-CNN、CenterNet等经典模型和YOLOv8s、YOLOv7n-tiny等先进模型相比，检测精度提高了1.0%~17.9%。模型经TensorRT加速引擎优化加速后，在NVIDIA Jetson TX2与NVIDIA Jetson Nano平台上的检测速率提高近1倍，同时不影响检测精度。

Abstract

To address the difficulty in fully identifying pavement defects of different sizes caused by insufficient multi-scale feature extraction capability of existing networks

a pavement disease detection model based on multi-scale feature reinforcement was proposed in this paper. Firstly

a fast spatial pyramid pooling module based on mixed dilated convolution was constructed

and by stacking dilated convolutions with different dilation coefficients

the network receptive field was further expanded to capture a larger range of contextual information and preserve more spatial information. Next

a multi-path feature fusion network was designed to achieve cross level feature capture and reduce information loss during the feature fusion process through multiple branches and skip connections. The

-means clustering algorithm was used together with the Intersection over Union to obtain reasonable anchor boxes. In addition

a penalty term for area was designed in the loss function and a descent gradient was set up to improve the accuracy and efficiency of the predicted box regression. Finally

efficient attention through cross channel interaction was introduced to achieve interaction between important channels in the model. Experimental results show that in terms of detection accuracy

the proposed model was 4.0% higher than the original model YOLOv5s and 1.0% to 17.9% higher than classical models such as Faster R-CNN and CenterNet and advanced models such as YOLOv8s and YOLOv7n-tiny. After optimization with TensorRT acceleration engine

the detection speed on NVIDIA Jetson TX2 and NVIDIA Jetson Nano embedded platforms nearly doubled without compromising the detection accuracy.

关键词

Keywords

references

何铁军 , 李华恩 . 基于改进YOLOv5的路面病害检测模型 [J ] . 土木工程学报 , 2024 , 57 ( 2 ): 96 - 106 .

HE Tiejun , LI Huaen . The pavement damage detection model based on improved YOLOv5 [J ] . China Civil Engineering Journal , 2024 , 57 ( 2 ): 96 - 106 .

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 Aimin , 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 .

张跃飞 , 王敬飞 , 陈斌 , 等 . 基于改进的Mask R-CNN的公路裂缝检测算法 [J ] . 计算机应用 , 2020 , 40 ( S2 ): 162 - 165 .

ZHANG Yuefei , WANG Jingfei , CHEN Bin , et al . Pavement crack detection algorithm based on improved Mask R-CNN [J ] . Journal of Computer Applications , 2020 , 40 ( S2 ): 162 - 165 .

TANG Jinshan , GU Yanliang . Automatic crack detection and segmentation using a hybrid algorithm for road distress analysis [C ] // 2013 IEEE International Conference on Systems, Man, and Cybernetics . Piscataway, NJ, USA : IEEE , 2013 : 3026 - 3030 .

ZHAO Huili , QIN Guofeng , WANG Xingjian . Improvement of canny algorithm based on pavement edge detection [C ] // Proceedings of the 2010 3rd International Congress on Image and Signal Processing . Piscataway, NJ, USA : IEEE , 2010 : 964 - 967 .

LECUN Y , BENGIO Y , HINTON G . Deep learning [J ] . Nature , 2015 , 521 ( 7553 ): 436 - 444 .

ZHANG Lei , YANG Fan , ZHANG Yimin , et al . Road crack detection using deep convolutional neural network [C ] // 2016 IEEE International Conference on Image Processing (ICIP) . Piscataway, NJ, USA : IEEE , 2016 : 3708 - 3712 .

张伟光 , 钟靖涛 , 于建新 , 等 . 基于机器学习和图像处理的路面裂缝检测技术研究 [J ] . 中南大学学报(自然科学版) , 2021 , 52 ( 7 ): 2402 - 2415 .

ZHANG Weiguang , ZHONG Jingtao , YU Jianxin , et al . Research on pavement crack detection technology based on convolution neural network [J ] . Journal of Central South University (Science and Technology) , 2021 , 52 ( 7 ): 2402 - 2415 .

王丽苹 , 高瑞贞 , 张京军 , 等 . 基于卷积神经网络的混凝土路面裂缝检测 [J ] . 计算机科学 , 2019 , 46 ( S2 ): 584 - 589 .

WANG Liping , GAO Ruizhen , ZHANG Jingjun , et al . Crack detection of concrete pavement based on convolutional neural network [J ] . Computer Science , 2019 , 46 ( S2 ): 584 - 589 .

孙朝云 , 裴莉莉 , 李伟 , 等 . 基于改进Faster R-CNN的路面灌封裂缝检测方法 [J ] . 华南理工大学学报(自然科学版) , 2020 , 48 ( 2 ): 84 - 93 .

SUN Zhaoyun , PEI Lili , LI Wei , et al . Pavement sealed crack detection method based on improved Faster R-CNN [J ] . Journal of South China University of Technology (Natural Science Edition) , 2020 , 48 ( 2 ): 84 - 93 .

许贵阳 , 张诗泉 , 白堂博 . 基于改进Faster R-CNN的CRTSⅡ型轨道板裂缝检测方法 [J ] . 中国铁道科学 , 2023 , 44 ( 1 ): 106 - 113 .

XU Guiyang , ZHANG Shiquan , BAI Tangbo . Crack detection method of CRTSⅡ track slab based on Faster R-CNN improvement [J ] . China Railway Science , 2023 , 44 ( 1 ): 106 - 113 .

罗晖 , 贾晨 , 李健 . 基于改进YOLOv4的公路路面病害检测算法 [J ] . 激光与光电子学进展 , 2021 , 58 ( 14 ): 328 - 336 .

LUO Hui , JIA Chen , LI Jian . Road surface disease detection algorithm based on improved YOLOv4 [J ] . Laser & Optoelectronics Progress , 2021 , 58 ( 14 ): 328 - 336 .

ROY A M , BHADURI J . DenseSPH-YOLOv5: an automated damage detection model based on DenseNet and swin-transformer prediction head-enabled YOLOv5 with attention mechanism [J ] . Advanced Engineering Informatics , 2023 , 56 : 102007 .

LI Mingjue , WANG Hao , PENG Yuhang , et al . Edge-to-client real-time road damage detection system based on Yolov5 [C ] // 2021 China Automation Congress (CAC) . Piscataway, NJ, USA : IEEE , 2021 : 1221 - 1226 .

周中 , 闫龙宾 , 张俊杰 , 等 . 基于YOLOX-G算法的隧道裂缝实时检测 [J ] . 铁道科学与工程学报 , 2023 , 20 ( 7 ): 2751 - 2762 .

ZHOU Zhong , YAN Longbin , ZHANG Junjie , et al . Real-time detection of tunnel cracks based on YOLOX-G algorithm [J ] . Journal of Railway Science and Engineering , 2023 , 20 ( 7 ): 2751 - 2762 .

XIANG Wanni , WANG Haichen , XU Yuan , et al . Road disease detection algorithm based on YOLOv5s-DSG [J ] . Journal of Real-Time Image Processing , 2023 , 20 ( 3 ): 56 .

ALFARRARJEH A , TRIVEDI D , KIM S H , et al . A deep learning approach for road damage detection from smartphone images [C ] // 2018 IEEE International Conference on Big Data (Big Data) . Piscataway, NJ, USA : IEEE , 2018 : 5201 - 5204 .

MANDAL V , MUSSAH A R , ADU-GYAMFI Y . Deep learning frameworks for pavement distress classification: a comparative analysis [C ] // 2020 IEEE International Conference on Big Data (Big Data) . Piscataway, NJ, USA : IEEE , 2020 : 5577 - 5583 .

张凡 , 张鹏超 , 王磊 , 等 . 基于YOLOv5s的轻量化朱鹮检测算法研究 [J ] . 西安交通大学学报 , 2023 , 57 ( 1 ): 110 - 121 .

ZHANG Fan , ZHANG Pengchao , WANG Lei , et al . Research on lightweight crested ibis detection algorithm based on YOLOv5s [J ] . Journal of Xi'an Jiaotong University , 2023 , 57 ( 1 ): 110 - 121 .

王保宪 , 白少雄 , 赵维刚 . 基于特征增强学习的路面裂缝病害视觉检测方法 [J ] . 铁道科学与工程学报 , 2022 , 19 ( 7 ): 1927 - 1935 .

WANG Baoxian , BAI Shaoxiong , ZHAO Weigang . Pavement crack damage visual detection method based on feature reinforcement learning [J ] . Journal of Railway Science and Engineering , 2022 , 19 ( 7 ): 1927 - 1935 .

彭雨诺 , 刘敏 , 万智 , 等 . 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 .

PENG Yunuo , LIU Min , WAN Zhi , et al . A dual deep net work based on the improved YOLO for fast bridge surface defect detection [J ] . Acta Automatica Sinica , 2022 , 48 ( 4 ): 1018 - 1032 .

邹俊志 , 杨建喜 , 李昊 , 等 . 复杂背景下基于改进YOLO v3算法的桥梁表观病害识别 [J ] . 铁道科学与工程学报 , 2021 , 18 ( 12 ): 3257 - 3266 .

ZOU Junzhi , YANG Jianxi , LI Hao , et al . Bridge apparent damage detection based on the improved YOLO v3 in complex background [J ] . Journal of Railway Science and Engineering , 2021 , 18 ( 12 ): 3257 - 3266 .

ZHENG Zhaohui , WANG Ping , REN Dongwei , et al . Enhancing geometric factors in model learning and inference for object detection and instance segmentation [J ] . IEEE Transactions on Cybernetics , 2022 , 52 ( 8 ): 8574 - 8586 .

WANG Qilong , WU Banggu , ZHU Pengfei , et al . ECA-net: efficient channel attention for deep convolutional neural networks [C ] // 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway, NJ, USA : IEEE , 2020 : 11531 - 11539 .

ARYA D , MAEDA H , GHOSH S K , et al . RDD2020: an annotated image dataset for automatic road damage detection using deep learning [J ] . Data in Brief , 2021 , 36 : 107133 .

RIBANI R , MARENGONI M . A survey of transfer learning for convolutional neural networks [C ] // 2019 32nd SIBGRAPI Conference on Graphics, Patterns and Images Tutorials (SIBGRAPI-T) . Piscataway, NJ, USA : IEEE , 2019 : 47 - 57 .

EVERINGHAM M , VAN GOOL L , WILLIAMS C K I , et al . The PASCAL visual object classes challenge 2007 [EB/OL ] . ( 2007-12-20 ) [ 2024-07-25 ] . http://host.robots.ox.ac.uk/pascal/VOC/voc2007/ http://host.robots.ox.ac.uk/pascal/VOC/voc2007/ .

NGUYEN K H , NGUYEN H V N , TRAN H N , et al . Combining autoencoder and Yolov6 model for classification and disease detection in chickens [C ] // Proceedings of the 2023 8th International Conference on Intelligent Information Technology . New York, NY, USA : Association for Computing Machinery , 2023 : 132 - 138 .

OKRAN A M , ABDEL-NASSER M , RASHWAN H A , et al . Effective deep learning-based ensemble model for road crack detection [C ] // 2022 IEEE International Conference on Big Data (Big Data) . Piscataway, NJ, USA : IEEE , 2022 : 6407 - 6415 .

倪昌双 , 李林 , 罗文婷 , 等 . 改进YOLOv7的沥青路面病害检测 [J ] . 计算机工程与应用 , 2023 , 59 ( 13 ): 305 - 316 .

NI Changshuang , LI Lin , LUO Wenting , et al . Disease detection of asphalt pavement based on improved YOLOv7 [J ] . Computer Engineering and Applications , 2023 , 59 ( 13 ): 305 - 316 .

TAN Mingxing , PANG Ruoming , LE Q V . EfficientDet: scalable and efficient object detection [C ] // 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway, NJ, USA : IEEE , 2020 : 10778 - 10787 .

KLUGER F , REINDERS C , RAETZ K , et al . Region-based cycle-consistent data augmentation for object detection [C ] // 2018 IEEE International Conference on Big Data (Big Data) . Piscataway, NJ, USA : IEEE , 2018 : 5205 - 5211 .

WANG Wenzhe , WU Bin , YANG Sixiong , et al . Road damage detection and classification with faster R-CNN [C ] // 2018 IEEE International Conference on Big Data (Big Data) . Piscataway, NJ, USA : IEEE , 2018 : 5220 - 5223 .

晏班夫 , 徐观亚 , 栾健 , 等 . 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 .

YAN Banfu , XU Guanya , LUAN Jian , et al . Pavement distress detection based on Faster R-CNN and morphological operations [J ] . China Journal of Highway and Transport , 2021 , 34 ( 9 ): 181 - 193 .

GE Zheng , LIU Songtao , WANG Feng , et al . YOLOX: exceeding YOLO series in 2021 [EB/OL ] . ( 2021-08-06 ) [ 2024-06-01 ] . https://arxiv.org/abs/2107.08430 https://arxiv.org/abs/2107.08430 .

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 .

WANG Yanbo , DING Ming , KAN Shichao , et al . Deep proposal and detection networks for road damage detection and classification [C ] // 2018 IEEE International Conference on Big Data (Big Data) . Piscataway, NJ, USA : IEEE , 2018 : 5224 - 5227 .

DUAN Kaiwen , BAI Song , XIE Lingxi , et al . CenterNet: keypoint triplets for object detection [C ] // 2019 IEEE/CVF International Conference on Computer Vision (ICCV) . Piscataway, NJ, USA : IEEE , 2019 : 6568 - 6577 .

郑美俊 , 田益民 , 杨帅 . 基于改进FCOS网络的遥感目标检测 [J ] . 航天返回与遥感 , 2022 , 43 ( 5 ): 133 - 141 .

ZHENG Meijun , TIAN Yimin , YANG Shuai . Remote sensing target detection based on improved FCOS network [J ] . Spacecraft Recovery & Remote Sensing , 2022 , 43 ( 5 ): 133 - 141 .

ANGULO A , VEGA-FERNÁNDEZ J A , AGUILAR-LOBO L M , et al . Road damage detection acquisition system based on deep neural networks for physical asset management [C ] // Advances in Soft Computing . Cham : Springer International Publishing , 2019 : 3 - 14 .

LI Xiang , LI Li , LIU Zhigui , et al . Dam crack detection studies by UAV based on YOLO algorithm [C ] // 2023 2nd International Conference on Robotics, Artificial Intelligence and Intelligent Control (RAIIC) . Piscataway, NJ, USA : IEEE , 2023 : 104 - 108 .

顾德英 , 罗聿伦 , 李文超 . 基于改进YOLOv5算法的复杂场景交通目标检测 [J ] . 东北大学学报(自然科学版) , 2022 , 43 ( 8 ): 1073 - 1079 .

GU Deying , LUO Yulun , LI Wenchao . Traffic target detection in complex scenes based on improved YOLOv5 algorithm [J ] . Journal of Northeastern University (Natural Science) , 2022 , 43 ( 8 ): 1073 - 1079 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

采用机器学习的大规模并行程序I/O性能建模与预测方法

水蒸气掺混对氨裂解气旋流燃烧与排放特性的影响研究

语义区分和子图整合驱动的知识图谱表示学习方法

电动车辆间接热系统快速除湿响应迟滞机理及优化控制策略

超临界二氧化碳向心透平轮背间隙内风阻损失及流动特性研究