Research on Improved Pavement Distress Detection Algorithm of Yolov12
Journal: Modern Economics & Management Forum DOI: 10.32629/memf.v6i5.4524
Abstract
Pavement health significantly affects traffic safety, regional economic growth, and residents' living standards. To address the inefficiency and inaccuracy of traditional manual inspection methods, this paper proposes a pavement distress detection algorithm based on an improved YOLOv12 model. By incorporating DySample upsampling and the NWD (Normalized Wasserstein Distance) Loss function, the algorithm enhances detection accuracy and robustness in complex scenarios. Validated on the UAV-PDD2023 dataset, the improved YOLOv12-ND model achieves 83.9% mAP50 and 58.4% mAP50-95, outperforming mainstream algorithms like YOLOv11, YOLOv10, and YOLOv9, especially in detecting repairs and longitudinal cracks. This study offers a scientific basis for pavement maintenance decisions and advances intelligent, precise pavement detection technology, crucial for building a safe and efficient transportation system.
Keywords
pavement distress detection; YOLOv12; DySample upsampling; NWD Loss; deep learning
Full Text
PDF - Viewed/Downloaded: 2 TimesReferences
[1] Cao W, Liu Q, He Z. Review of pavement defect detection methods[J]. IEEE access, 2020, 8: 14531-14544.
[2] Cao W, Liu Q, He Z. Review of pavement defect detection methods[J]. IEEE access, 2020, 8: 14531-14544.
[3] 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.
[4] 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.
[5] Guo, G., Zhang, Z. Road damage detection algorithm for improved YOLOv5. Sci Rep 12, 15523 (2022). https://doi.org/10.1038/s41598-022-19674-8
[6] Ren, M., Zhang, X., Zhi, X. et al. An annotated street view image dataset for automated road damage detection. Sci Data 11, 407 (2024). https://doi.org/10.1038/s41597-024-03263-7
[7] Pang, D., Guan, Z., Luo, T. et al. Real-time detection of road manhole covers with a deep learning model. Sci Rep 13, 16479 (2023). https://doi.org/10.1038/s41598-023-43173-z
[8] Yan H, Zhang J. UAV-PDD2023: A benchmark dataset for pavement distress detection based on UAV images[J]. Data in Brief, 2023, 51: 109692.
[9] Liu W, Lu H, Fu H, et al. Learning to upsample by learning to sample[C]//Proceedings of the IEEE/CVF international conference on computer vision. 2023: 6027-6037.
[10] Wang J, Xu C, Yang W, et al. A normalized Gaussian Wasserstein distance for tiny object detection[J]. arXiv preprint arXiv:2110.13389, 2021.
[11] Tian Y, Ye Q, Doermann D. Yolov12: Attention-centric real-time object detectors[J]. arXiv preprint arXiv:2502.12524, 2025.
[12] Dao T, Fu D, Ermon S, et al. Flashattention: Fast and memory-efficient exact attention with io-awareness[J]. Advances in neural information processing systems, 2022, 35: 16344-16359.
[2] Cao W, Liu Q, He Z. Review of pavement defect detection methods[J]. IEEE access, 2020, 8: 14531-14544.
[3] 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.
[4] 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.
[5] Guo, G., Zhang, Z. Road damage detection algorithm for improved YOLOv5. Sci Rep 12, 15523 (2022). https://doi.org/10.1038/s41598-022-19674-8
[6] Ren, M., Zhang, X., Zhi, X. et al. An annotated street view image dataset for automated road damage detection. Sci Data 11, 407 (2024). https://doi.org/10.1038/s41597-024-03263-7
[7] Pang, D., Guan, Z., Luo, T. et al. Real-time detection of road manhole covers with a deep learning model. Sci Rep 13, 16479 (2023). https://doi.org/10.1038/s41598-023-43173-z
[8] Yan H, Zhang J. UAV-PDD2023: A benchmark dataset for pavement distress detection based on UAV images[J]. Data in Brief, 2023, 51: 109692.
[9] Liu W, Lu H, Fu H, et al. Learning to upsample by learning to sample[C]//Proceedings of the IEEE/CVF international conference on computer vision. 2023: 6027-6037.
[10] Wang J, Xu C, Yang W, et al. A normalized Gaussian Wasserstein distance for tiny object detection[J]. arXiv preprint arXiv:2110.13389, 2021.
[11] Tian Y, Ye Q, Doermann D. Yolov12: Attention-centric real-time object detectors[J]. arXiv preprint arXiv:2502.12524, 2025.
[12] Dao T, Fu D, Ermon S, et al. Flashattention: Fast and memory-efficient exact attention with io-awareness[J]. Advances in neural information processing systems, 2022, 35: 16344-16359.
Copyright © 2025 Nan Deng, Quan Song
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License