面向航拍路面裂缝检测的AC-YOLO

doi:10.3778/j.issn.1002-8331.2407-0139

摘要/Abstract

摘要： 针对当前道路巡检智能化程度不足以及效率低等现状，为提高利用无人机道路巡检检测效率与检测精度，在YOLOv8s的基础上针对无人机航拍场景提出改进模型AC-YOLO，在主干网络引入动态大卷积核注意机制LSK-attention来扩展模型的感受野，提高对路面裂缝范围检测的准确性。在颈部结构设计多尺度特征融合策略，融入BiFPN网络，改善对细小裂缝的检测。替换损失函数为WIoUv3，优化梯度分配策略，使模型更加关注裂缝主体。在数据集UAV-PDD2023上进行实验验证，改进后AC-YOLO精准度达到0.895，较原模型提高0.128，mAP50达到0.791，提高0.071，F1得分提高0.051，模型大小减小8.5%，FPS达到了129，提高4%。同时采用不同任务验证了模型泛化性能，实验结果证明AC-YOLO具有更强的检测性能，能有效应用于无人机视角下的路面裂缝检测。

关键词: 道路巡检, 裂缝检测, 航拍裂缝, 增强感知, 特征融合

Abstract: In response to the current issues of insufficient automation and low efficiency in road inspection, this study aims to enhance the efficiency and accuracy of UAV-based road inspection. Based on YOLOv8s, an improved model, AC-YOLO, is proposed specifically for UAV aerial scenarios. First, the model integrates a dynamic large-kernel convolutional attention mechanism, LSK-attention, into the backbone network to expand the receptive field and improve the model’s accuracy in detecting road crack areas. Second, a multi-scale feature fusion strategy is introduced in the neck structure by incorporating the BiFPN network, enhancing the model’s ability to detect fine cracks. Finally, the loss function is replaced with WIoUv3, optimizing the gradient allocation strategy to enable the model to focus more precisely on crack regions. Experimental validation on the UAV-PDD2023 dataset demonstrates that the improved AC-YOLO achieves an accuracy of 0.895, representing a 0.128 increase compared to the original model. The mAP50 reaches 0.791, reflecting an improvement of 0.071, while the F1 score increases by 0.051. Moreover, the model’s size is reduced by 8.5%, and the FPS reaches 129, showing a 4% improvement. The model’s generalization performance is verified across multiple tasks, and the experimental results confirm that AC-YOLO offers superior detection capabilities, making it highly effective for UAV-based road crack detection.

Key words: road inspection, crack detection, aerial cracks, augmented perception, feature fusion

白锋, 马庆禄, 赵敏. 面向航拍路面裂缝检测的AC-YOLO[J]. 计算机工程与应用, 2025, 61(1): 153-164.

BAI Feng, MA Qinglu, ZHAO Min. AC-YOLO for Aerial Pavement Crack Detection[J]. Computer Engineering and Applications, 2025, 61(1): 153-164.

参考文献

[1] 赵志宏, 郝子晔. 改进YOLOv8的航拍小目标检测方法: CRP-YOLO[J]. 计算机工程与应用, 2024, 60(13): 209-218.
ZHAO Z H, HAO Z Y. Improved YOLOv8 aerial small target detection method: CRP-YOLO[J]. Computer Engineering and Applications, 2024, 60(13): 209-218.
[2] 初秀民, 严新平, 陈先桥. 路面破损图像二值化方法研究[J]. 计算机工程与应用, 2008, 44(28): 161-165.
CHU X M, YAN X P, CHEN X Q. Study of pavement surface distress image binarization[J]. Computer Engineering and Applications, 2008, 44(28): 161-165.
[3] 翟军治, 孙朝云, 裴莉莉, 等. 多尺度特征增强的路面裂缝检测方法[J]. 交通运输工程学报, 2023, 23(1): 291-308.
ZHAI J Z, SUN Z Y, PEI L L, et al. Pavement crack detection method based on multi-scale feature enhancement[J]. Journal of Traffic and Transportation Engineering, 2023, 23(1): 291-308.
[4] 陈涵深, 姚明海, 瞿心昱. 基于U型全卷积神经网络的路面裂缝检测[J]. 光电工程, 2020, 47(12): 67-77.
CHEN H S, YAO M H, QU X Y. Pavement crack detection based on the U-shaped fully convolutional neural network[J]. Opto-Electron Eng, 2020, 47(12): 67-77.
[5] 夏晓华, 苏建功, 王耀耀, 等. 基于DeepLabv3+的轻量化路面裂缝检测模型[J]. 激光与光电子学进展, 2024, 61(8): 172-181.
XIA X H, SU J G, WANG Y Y, et al. Lightweight pavement crack detection model based on DeepLabv3+[J]. Laser and Optoelectronics Progress, 2024, 61(8): 172-181.
[6] 肖力炀, 李伟, 袁博, 等. 一种基于改进实例分割模型的路面裂缝检测方法[J]. 武汉大学学报 (信息科学版), 2023, 48(5): 765-776.
XIAO L Y, LI W, YUAN B, et al. A pavement crack identification method based on improved instance segmentation model[J]. Geomatics and Information Science of Wuhan University, 2023, 48(5): 765-776.
[7] 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 T J, LI H E. The pavement damage detection model based on improved YOLOv5[J]. China Civil Engineering Journal, 2024, 57(2): 96-106.
[8] INAM H , ISLAM N U, AKRAM M U, et al. Smart and automated infrastructure management: a deep learning approach for crack detection in bridge images[J]. Sustainability, 2023, 15(3): 1866-1866.
[9] 李泽伟, 杨永清, 廖曼, 等. 基于计算机视觉与混合测量技术的结构裂缝识别方法[J/OL]. 西南交通大学学报: 1-12[2024-07?16].http://kns.cnki.net/kcms/detail/51.1277.U.20240712.
1344.002.html.
LI Z W, YANG Y Q, LIAO M, et al. Structural crack detection based on computer vision and hybrid measurement technology[J/OL]. Journal of Southwest Jiaotong University: 1-12[2024-07-16]. http://kns.cnki.net/kcms/detail/51.1277.U.
20240712.1344.002.html.
[10] WANG X, GAO H, JIA Z, et al. BL-YOLOv8: an improved road defect detection model based on YOLOv8[J]. Sensors, 2023, 23(20): 8361.
[11] XIONG C, ZAYED T, ABDELKADER E M. A novel YOLOv8-GAM-Wise-IoU model for automated detection of bridge surface cracks[J]. Construction and Building Materials, 2024, 414: 135025.
[12] 王雪秋, 高焕兵, 郏泽萌. 改进YOLOv8的道路缺陷检测算法[J]. 计算机工程与应用, 2024, 60(17): 179-190.
WANG X Q, GAO H B, JIA Z M. Improved road defect detection algorithm based on YOLOv8[J]. Computer Engineering and Applications, 2024, 60(17): 179-190.
[13] 胥铁峰, 黄河, 张红民, 等. 基于改进YOLOv8的轻量化道路病害检测方法[J]. 计算机工程与应用, 2024, 60(14): 175-186.
XU T F, HUANG H, ZHANG H M, et al. Lightweight road damage detection method based on improved YOLOv8[J]. Computer Engineering and Applications, 2024, 60(14): 175-186.
[14] QIU Q W, DENVID L. Real-time detection of cracks in tiled sidewalks using YOLO-based method applied to unmanned aerial vehicle (UAV) images[J]. Automation in Construction, 2023, 147: 104745.
[15] WANG G , CHEN Y, AN P, et al. UAV-YOLOv8: a small-object-detection model based on improved YOLOv8 for UAV aerial photography scenarios[J]. Sensors, 2023, 23(16): 7190.
[16] 程换新, 乔庆元, 骆晓玲, 等. 基于改进YOLOv8的无人机航拍图像目标检测算法[J]. 无线电工程, 2024, 54(4): 871-881.
CHENG H X, QIAO Q Y, LUO X L, et al. Object detection algorithm for UAV aerial image improved by YOLOv8[J]. Radio Engineering , 2024, 54(4): 871-881.
[17] ZHENG Z, WANG P, LIU W, et al. Distance-IoU loss: faster and better learning for bounding box regression[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2020: 12993-13000.
[18] FENG C, ZHONG Y, GAO Y, et al. Tood: task-aligned one-stage object detection[C]//Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision, 2021: 3490-3499.
[19] 李松, 史涛, 井方科. 改进YOLOv8的道路损伤检测算法[J]. 计算机工程与应用, 2023, 59(23): 165-174.
LI S, SHI T, JING F K. Improved road damage detection algorithm of YOLOv8[J]. Computer Engineering and Applications, 2023, 59(23): 165-174.
[20] LAU K W, PO L M, REHMAN Y A U. Large separable kernel attention: rethinking the large kernel attention design in CNN[J]. Expert Systems with Applications, 2024, 236: 121352.
[21] LI Y, HOU Q, ZHENG Z, et al. Large selective kernel network for remote sensing object detection[J]. arXiv:2303. 09030, 2023.
[22] TAN M, PANG R, LE Q V. EfficientDet: scalable and efficient object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 10781-10790.
[23] TONG Z, CHEN Y, XU Z, et al. Wise-IoU: bounding box regression loss with dynamic focusing mechanism[J]. arXiv:2301.10051, 2023.
[24] YAN H, ZHANG J. UAV-PDD2023: a benchmark dataset for pavement distress detection based on UAV images[J]. Data in Brief, 2023, 51: 109692.
[25] HONG Z H, YANG F, PAN H, et al. Highway crack segmentation from unmanned aerial vehicle images using deep learning[J]. IEEE Geoscience and Remote Sensing Letters, 2021, 19: 1-5.