Research on Pavement Anomaly Detection Technology Based on Improved RT-DETR

doi:10.19678/j.issn.1000-3428.0070182

Abstract

Abstract:

Pavement anomaly detection holds significant practical importance for ensuring driving safety, optimizing traffic management, and enhancing driving experience. To address the challenges posed by variations in the size, shape, and color of pavement anomalies, and complex environmental interferences that lower detection accuracy and efficiency, this study proposes an improved Real-Time Detection Transformer (RT-DETR)-based technology for pavement anomaly target detection. First, a Large Receptive Field Element-wise Multiplication Block (LRFEM_Block) is designed to replace the BasicBlock module in the original backbone network, effectively enhancing feature expression capabilities based on the element-wise multiplication principle. Next, a Generalized Efficient Layer Aggregation Network (GELAN) is introduced and combined with multi-scale LRFEM_Block modules to design a Multiplicative-based Layer Aggregation Intra-scale Feature Interaction (MLA-IFI) structure, which improves the computational efficiency and performance of the neck network for deep features and optimizes the gradient propagation path. Additionally, the Selective Boundary Aggregation (SBA) concept is employed to construct a Bidirectional Adaptive Boundary Fusion Feature Pyramid Network (BABF-FPN) multi-scale feature fusion module, adaptively aggregating features of different resolutions bidirectionally and promoting the refinement of small object boundaries. Experimental results show that the improved method achieves a 3.4 and 4.7 percentage point increase in mAP@0.5 on a self-built dataset and the RDD2022 public dataset, respectively, outperforming other models. Moreover, it reduces the number of parameters and computational load by 24.5% and 11.2%, respectively, with a detection speed of 74 frame/s, thereby satisfying the deployment requirements for in-vehicle pavement anomaly detection.

Key words: pavement anomaly detection, Real-Time Detection Transformer (RT-DETR) algorithm, element-wise multiplication, Generalized Efficient Layer Aggregation Network (GELAN) structure, recalibration

摘要：

路面异常检测对于保障行车安全、优化交通管理和驾乘体验具有重要的现实意义。针对路面异常物体在尺寸、形状及颜色等方面跨度较大, 以及复杂环境干扰导致检测精度和效率较低的问题, 提出一种改进实时检测Transformer(RT-DETR)的路面异常目标检测技术。首先, 设计大感受野元素乘法模块(LRFEM_Block)替代原主干网络中的BasicBlock模块, 该模块依据元素乘法原理有效增强特征表达能力。其次, 引入广义高效层聚合网络(GELAN)思想, 并结合多尺度LRFEM_Block设计一种基于元素乘法的层聚合尺度内特征交互(MLA-IFI)结构, 提高颈部网络对深层特征的计算效率和性能, 优化梯度传播路径。此外, 引入自适应选择边界聚合(SBA)思想构建双向自适应边界融合特征金字塔网络(BABF-FPN)多尺度特征融合模块, 自适应双向聚合不同分辨率特征, 促进小目标物体模糊边界的细粒度化。实验结果表明, 改进方法在自建数据集和RDD2022公开数据集上的mAP@0.5相较于基线算法分别提升3.4和4.7百分点, 均优于其他测试模型, 并且参数量和计算量分别减少24.5%和11.2%, 算法检测速度达到74帧/s, 更加契合车载路面异常检测的部署需求。

关键词: 路面异常检测, 实时检测Transformer算法, 元素乘法, 广义高效层聚合网络结构, 重校准

LIU Ze, SONG Tinglun, SHI Xianrang, SU Yang, ZHAO Qun. Research on Pavement Anomaly Detection Technology Based on Improved RT-DETR[J]. Computer Engineering, 2026, 52(4): 187-199.

刘泽, 宋廷伦, 石先让, 苏洋, 赵群. 基于改进RT-DETR的路面异常检测技术研究[J]. 计算机工程, 2026, 52(4): 187-199.

/ Recommend / Download Citations

URL: https://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0070182

https://www.ecice06.com/EN/Y2026/V52/I4/187

Figures/Tables 21

Fig.1 RT-DETR model structure

Fig.2 LRFEM_Block structure

Fig.3 MLA-IFI module structure

Fig.4 SBA module structure

Fig.5 RFU module structure

Fig.6 BABF-FPN structure

Fig.7 Improved algorithm model structure

Fig.8 Examples of different road surface anomalies in the dataset

Fig.9 Experimental comparison of visual heat maps

Fig.10 Normal environment visualization comparison experiment

Fig.11 Shadow environment visualization comparison experiment

Fig.12 Comparison of average detection precision

Fig.13 Visualization comparison experiment of loss functions

References 28

1	孔凡伟. 智能交通系统中路面质量对货运安全的影响分析与改进策略. 中国航务周刊, 2024 (3): 78- 80. URL
	KONG F . Analysis and improvement strategies of the impact of pavement quality on freight safety in intelligent transportation systems. China Shipping Gazette, 2024 (3): 78- 80. URL
2	RYU S , PARK Y , SUH M . Ride quality analysis of a tracked vehicle suspension with a preview control. Journal of Terramechanics, 2011, 48 (6): 409- 417. doi: 10.1016/j.jterra.2011.09.002
3	魏明明. 基于车载3D加速传感器的路况监测研究[D]. 大连: 大连理工大学, 2011.
	WEI M. Research on road condition monitoring based on in-vehicle 3D accelerometers[D]. Dalian: Dalian University of Technology, 2011. (in Chinese)
4	LI Z J , KOLMANOVSKY I , ATKINS E , et al. H∞ filtering for cloud-aided semi-active suspension with delayed road information. IFAC-PapersOnLine, 2015, 48 (12): 275- 280. doi: 10.1016/j.ifacol.2015.09.390
5	潘勇. 基于图像三维重建的路面裂缝检测研究[D]. 北京: 北京建筑大学, 2023.
	PAN Y. Research on road surface crack detection based on image-based 3D reconstruction[D]. Beijing: Beijing University of Civil Engineering and Architecture, 2023. (in Chinese)
6	孟波, 史旭华, 张彬. 基于双分支卷积和深度插值的点云表面重建. 计算机工程, 2025, 51 (7): 119- 126. doi: 10.19678/j.issn.1000-3428.0069420
	MENG B , SHI X H , ZHANG B . Point cloud surface reconstruction based on dual-branch convolution and depth interpolation. Computer Engineering, 2025, 51 (7): 119- 126. doi: 10.19678/j.issn.1000-3428.0069420
7	MENANT F, MARTIN J M, RADDE B, et al. A surface-based analysis applied to road deformation: Methods developed and experimental results[C]//Proceedings of the 27th IEEE International Conference on World Road Congress. Washington D. C., USA: IEEE Press, 2023: 437-446.
8	TSAI Y J , CHATTERJEE A . Pothole detection and classification using 3D technology and watershed method. Journal of Computing in Civil Engineering, 2018, 32 (2): 04017078. doi: 10.1061/(ASCE)CP.1943-5487.0000726
9	CORD A , CHAMBON S . Automatic road defect detection by textural pattern recognition based on AdaBoost. Computer-Aided Civil and Infrastructure Engineering, 2012, 27 (4): 244- 259. doi: 10.1111/j.1467-8667.2011.00736.x
10	OUMA Y O , HAHN M . Pothole detection on asphalt pavements from 2D-colour pothole images using fuzzy c-means clustering and morphological reconstruction. Automation in Construction, 2017, 83, 196- 211. doi: 10.1016/j.autcon.2017.08.017
11	MA N C , FAN J H , WANG W S , et al. Computer vision for road imaging and pothole detection: a state-of-the-art review of systems and algorithms. Transportation Safety and Environment, 2022, 4 (4): 45- 60.
12	GOU C, PENG B, LI T R, et al. Pavement crack detection based on the improved faster-RCNN[C]//Proceedings of the 14th IEEE International Conference on Intelligent Systems and Knowledge Engineering. Washington D. C., USA: IEEE Press, 2019: 962-967.
13	LI S B, HUANG Y J. Damage detection algorithm based on faster-RCNN[C]//Proceedings of the 5th International Conference on Electronics and Communication, Network and Computer Technology. Washington D. C., USA: IEEE Press, 2023: 177-180.
14	马荣贵, 黄训燕, 董世浩. 基于改进YOLOv8的实时坑槽检测算法研究. 计算机工程, 2025, 51 (11): 226- 234. doi: 10.19678/j.issn.1000-3428.0069671
	MA R G , HUANG X Y , DONG S H . Research on real-time pothole detection algorithm based on improved YOLOv8. Computer Engineering, 2025, 51 (11): 226- 234. doi: 10.19678/j.issn.1000-3428.0069671
15	WANG C Y, YEH I H, LIAO H Y. YOLOv9: learning what you want to learn using programmable gradient information[EB/OL]. [2024-06-20]. https://arxiv.org/abs/2402.13616.
16	GILG J, TEEPE T, HERZOG F, et al. Do we still need non-maximum suppression? Accurate confidence estimates and implicit duplication modeling with IoU-aware calibration[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. Waikoloa, USA: IEEE Press, 2024: 4850-4859.
17	ZHAO Y A, LÜ W Y, XU S L, et al. DETRs beat YOLOs on real-time object detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle, USA: IEEE Press, 2024: 16965-16974.
18	ARYA D , MAEDA H , GHOSH S K , et al. RDD2022: a multi-national image dataset for automatic road damage detection. Geoscience Data Journal, 2024, 3, 260- 271.
19	VASWANI A, SHAZWWR N, PARMAR N, et al. Attention is all you need[C]//Proceedings of the Advances in Neural Information Processing Systems. Cambridge, USA: MIT Press, 2017: 303-312.
20	HAN K , WANG Y H , CHEN H T , et al. A survey on vision transformer. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45 (1): 87- 110. doi: 10.1109/TPAMI.2022.3152247
21	CARION N, MASSA F, SYNNAEVE G, et al. End-to-end object detection with transformers[C]// Proceedings of the European Conference on Computer Vision. Berlin, Germany: Springer, 2020: 213-229.
22	LI F, ZENG A L, LIU S L, et al. Lite DETR: an interleaved multi-scale encoder for efficient DETR[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2023: 18558-18567.
23	SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[EB/OL]. [2024-06-20]. https://arxiv.org/abs/1409.1556.
24	HASTIE T , TIBSHIRANI R , FRIEDMAN J H , et al. The elements of statistical learning: data mining, inference, and prediction. New York, USA: ACM Press, 2009.
25	MA X, DAI X Y, BAI Y, et al. Rewrite the stars[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2024: 5694-5703.
26	TANG F, XU Z, HUANG Q, et al. DuAT: dual-aggregation transformer network for medical image segmentation[C]//Proceedings of Conference on Pattern Recognition and Computer Vision. Berlin, Germany: Springer, 2023: 343-356.
27	NIENABER S, BOOYSEN M J, KROON R S. Detecting potholes using simple image processing techniques and real-world footage[C]//Proceedings of the South African Transport Conference. South African, Pretoria: University of Pretoria Press, 2015: 422-434.
28	WANG Z, LI C, XU H, et al. Mamba YOLO: SSMs-based YOLO for object detection[EB/OL]. [2024-06-20]. https://arxiv.org/abs/2406.05835.

Please choose a citation manager

Content to export