Cracks are a typical distress in airport runway pavements. Timely detection is crucial for enhancing pavement performance and preventing their progression into structural damage. To address the current issues of insufficient detection accuracy under low illumination and noise interference, as well as limited ability to identify cracks of different scales in runway crack detection, an improved AER-UNet detection model was proposed. Three aspects of the proposed method were improved based on the UNet model. Firstly, an ASPP module was integrated at the end of the encoder to enhance the capability for multi-scale crack feature extraction. Secondly, an ECA attention mechanism was embedded into skip connections to improve robustness against noise and low-contrast backgrounds. Finally, the standard convolutions of the UNet model were replaced with residual convolution blocks to alleviate the vanishing gradient problem. In addition, a crack parameter quantification algorithm was proposed to address the limitation of existing methods that only provide segmentation results. Comparative experiments and visualization analyses demonstrate that the performance of AER-UNet was superior to that of the original UNet and other comparative models. Its precision, recall, and F1 score reach 85.7%, 84.5%, and 85.1%, respectively. Furthermore, crack parameter quantification algorithm possesses high accuracy, which can provide data support for the airport runway crack damage assessment and scientific maintenance decision-making.
LIANG Jin-tao, LI Ming-guang, MAO Wei, CHEN Jin-jian, LI Chao-feng, HOU Pu-min.
Crack Segmentation and Parameter Calculation for Airport Runways Based on AER-UNet[J].
Science Technology and Engineering, 2026, 26(14): 6108-6116 DOI:10.12404/j.issn.1671-1815.2506151
UNet 是一种经典的语义分割网络,其架构呈对称的 U 形结构,主要由编码器、解码器以及连接二者之间的跳跃连接三部分构成。编码器通过多级下采样操作逐步提取图像的高层语义特征,而解码器则通过相应的上采样过程逐步恢复空间分辨率;在此过程中,跳跃连接将编码器各层级所提取的浅层细节信息与解码器中对应的深层语义特征进行融合。然而,在面对复杂背景干扰(如光照、噪声干扰)以及不同尺度裂缝时,原始UNet架构在特征表达能力与信息感知方面仍显不足,存在一定的局限性。
KheradmandiN, MehranfarV. A critical review and comparative study on image segmentation-based techniques for pavement crack detection[J]. Construction and Building Materials, 2022, 321. DOI: 10.1016/j.conbuildmat.2021.126162.
[2]
TangW, WuZ, WangW, et al. VM-UNet++ research on crack image segmentation based on improved VM-UNet[J]. Scientific Reports, 2025, 15(1). DOI: 10.1038/s41598-025-92994-7.
WangAnzheng, DangJianwu, YueBiao, et al. Pavement crack detection based on positional information and attention mechanism[J]. Computer Engineering, 2024, 50(4): 303-312.
DuLei, ChenXi, BaiPengpeng, et al. Asphalt pavement crack detection method based on improved YOLOv5[J]. Science Technology and Engineering, 2024, 24(31): 13579-13591.
[9]
ZhangH, MaL, YuanZ, et al. Enhanced concrete crack detection and proactive safety warning based on I-ST-UNet model[J]. Automation in Construction, 2024, 166. DOI: 10.1016/j.autcon.2024.105612.
[10]
SafaeiN, SmadiO, MasoudA, et al. An automatic image processing algorithm based on crack pixel density for pavement crack detection and classification[J]. International Journal of Pavement Research and Technology, 2022, 15(1): 159-172.
[11]
LuoJ, LinH, WeiX, et al. Adaptive canny and semantic segmentation networks based on feature fusion for road crack detection[J]. IEEE Access, 2023, 11: 51740-51753.
[12]
ChenX, LiJ, HuangS, et al. An automatic concrete crack-detection method fusing point clouds and images based on improved Otsu’s algorithm[J]. Sensors, 2021, 21(5). DOI: 10.3390/s2105158.
ZhaiJunzhi, SunChaoyun, PeiLili, et al. Multi-scale feature enhancement method for pavement crack detection[J]. Journal of Traffic and Transportation Engineering, 2023, 23(1): 291-308.
[15]
ShiM, LiH, YaoQ, et al. Vision based nighttime pavement cracks pixel level detection by integrating infrared visible fusion and deep learning[J]. Construction and Building Materials, 2024, 442. DOI: 10.1016/j.conbuildmat.2024.137662.
[16]
WangJ, ShenB, LiG, et al. A novel multi-scale feature enhancement U-shaped network for pixel-level road crack segmentation[J]. Electronics, 2024, 13(22). DOI: 10.3390/electronics13224503.
[17]
YuG, DongJ, WangY, et al. RUC-Net: a residual-UNet-based convolutional neural network for pixel-level pavement crack segmentation[J]. Sensors, 2023, 23(1). DOI: 10.3390/s23010053.
MaYafei, SunWenkang, HeYu, et al. Concrete bridge surface crack recognition method based on DC-Unet[J]. Journal of Chang’an University (Natural Science Edition), 2024, 44(3): 66-75.
[20]
WuQ, SongZ, ChenH, et al. A highway pavement crack identification method based on an improved U-Net model[J]. Applied Sciences, 2023, 13(12). DOI: 10.3390/app13127227.
LuGuanhong, LüChengshun, TianJuan, et al. Highway tunnel lining crack detection based on improved YOLOv5[J]. Science Technology and Engineering, 2025, 25(7): 2997-3006.
[23]
RedmonJ, DivvalaS, GirshickR, et al. You only look once: unified, real-time object detection[C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE, 2016: 779-788.
[24]
GirshickR, DonahueJ, DarrellT, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Columbus: IEEE, 2014: 580-587.
ZhangYing, WangJixu, CaoYingkang, et al. A road mixed disease algorithm based on improved YOLOv9-c[J]. Science Technology and Engineering, 2025, 25(18): 7793-7802.
ChengQing, YanHengzhi, LiYandong. A lightweight small UAV detection method based on improved YOLOv8n[J]. Science Technology and Engineering, 2025, 25(28): 12098-12107.
ZhaiYongjie, ZhuHaoshuo, LiJiawei, et al. VDD-YOLO: an assessment method for vehicle deformation damage[J]. Science Technology and Engineering, 2025, 25(29): 12560-12569.
[35]
LongJ, ShelhamerE, DarrellT. Fully convolutional networks for semantic segmentation[C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Boston: IEEE, 2015: 3431-3440.
[36]
RonnebergerO, FischerP, BroxT. U-Net: convolutional networks for biomedical image segmentation[C]//Medical Image Computing and Computer-Assisted Intervention (MICCAI). Munich: Springer, 2015: 234-241.
[37]
LiuZ, CaoY, WangY, et al. Computer vision-based concrete crack detection using U-net fully convolutional networks[J]. Automation in Construction, 2019, 104: 129-139.
[38]
YangX, LiH, YuY, 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.
ShiJ, DangJ, CuiM, et al. Improvement of damage segmentation based on pixel-level data balance using VGG-UNet[J]. Applied Sciences, 2021, 11(2). DOI: 10.3390/app11020518.
[41]
LiB, XuH, JinX, et al. An optimized dual-view snake U-net model for tunnel lining crack detection[J]. Buildings, 2025, 15(5). DOI: 10.3390/buildings15050777.
[42]
GuoF, LiuJ, LüC, et al. A novel transformer-based network with attention mechanism for automatic pavement crack detection[J]. Construction and Building Materials, 2023, 391. DOI: 10.1016/j.conbuildmat.2023.131852.
DengZhilin, LuoRenze, FeiYue, et al. Airport pavement crack detection based on FE-UNet[J]. Journal of Optoelectronics·Laser, 2023, 34(1): 34-42.
[45]
WangQ, WuB, ZhuP, et al. ECA-Net: efficient channel attention for deep convolutional neural networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE, 2020: 11534-11542.