Aiming at the current issues in lane detection algorithms based on deep learning, such as insufficient real-time performance, limited global feature modeling capability, and the fact that most related studies remain at the simulation validation stage with a lack of deployment verification for practical embedded systems, this paper proposed a lane detection method based on a lightweight YOLO network and an attention mechanism, and completed the deployment and verification from algorithm design to a real-vehicle system. First, a lightweight Faster-Net was introduced as the backbone network, combined with partial convolution (PConv) modules for structural redesign, which significantly reduced computational complexity while maintaining feature extraction capability. Second, a self-attention mechanism was embedded after the SPPF module of the feature fusion network to enhance the model’s modeling ability for the global structure of lane lines and long-range spatial dependencies, while suppressing interference from complex backgrounds. The results show that the proposed method achieves a precision of 96.83% on the CULane dataset with a single-frame inference time of 11.0 ms, and an accuracy of 96.33% on the TuSimple dataset with a frame rate of 90.9 FPS, outperforming current mainstream algorithms. Finally, the optimized model was deployed on the Jetson Orin NX embedded platform, and functional verification was carried out in a real-vehicle sandbox environment. The results demonstrate that the system exhibits good stability and real-time performance in real-world scenarios. This work not only improves the theoretical performance of lane detection algorithms, but also transitions the technology from dataset validation to practical in-vehicle system implementation, providing a complete and feasible technical pathway for the engineering application of autonomous driving perception technology.
XUMingxing, WANGKunjun, LIUYang, et al. Status and prospects of synergistic development between smart city infrastructure and intelligent connected vehicles[J]. Journal of Tsinghua University (Science and Technology), 2025, 65(12): 2464-2492. (in Chinese)
LUOTongqiang, LIUJianjian, ZHAOBinggen, et al. Current status and future outlook of system functional safety mechanisms in intelligent vehicles[J]. Chinese Journal of Automotive Engineering, 2024, 14(6): 921-933. (in Chinese)
LUOWeiming, YINGZhaoyang, SUNWei, et al. Research on safety technological inspection requirements of vehicles with driving assist[J]. China Standardization, 2023(21): 219-223. (in Chinese)
ZHAOQiang, WANGRui, ZHUBaoquan, et al. Overview of the research progress of lane line detection based on machine vision[J]. Computing Technology and Automation, 2022, 41(1): 34-40. (in Chinese)
ZHOUFahua, CHENJiqing, YANGRong. Lane line recognition based on improved Hough transformation and slope feature[J]. Modern Electronics Technique, 2023, 46(7): 180-186. (in Chinese)
WANGJinnan, XUEChenyang, QIXiangdong, et al. Research on UUV trajectory tracking control based on RBF neural network PID[J]. Journal of North University of China (Natural Science Edition), 2024, 45(6): 843-851. (in Chinese)
[15]
SONGY, FENGQ, XIAOS, et al. Fault diagnosis of railway switch machines based on VMD⁃SDP-CNN[J]. Journal of Measurement Science and Instrumentation, 2025, 16(2): 291-301.
WULi, ZHANGZhenghao, GECaicheng, et al. Lane line detection algorithm based on improved SCNN network[J]. Computer and Modernization, 2024(7): 87-92. (in Chinese)
[18]
KRIZHEVSKYA, SUTSKEVERI, HINTONG E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90.
[19]
LIUB, LINGQ. Sparse LaneFormer: End-to-end lane detection with sparse proposals and interactions[J]. IEEE Transactions on Intelligent Transportation Systems, 2025, 26(8): 11931-11944.
[20]
CHENL, SIMAC, LIY, et al. PersFormer: 3D lane detection via perspective transformer and the OpenLane benchmark[C]//Computer Vision-ECCV 2022. Cham: Springer, 2022: 550-567.
[21]
BAIY, CHENZ, FUZ, et al. CurveFormer: 3D lane detection by curve propagation with curve queries and attention[C]//2023 IEEE International Conference on Robotics and Automation (ICRA), 2023: 7062-7068.
[22]
LIUR, CHEND, LIUT, et al. Learning to predict 3D lane shape and camera pose from a single image via geometry constraints[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2022, 36(2): 1765-1772.
[23]
HUANGS, SHENZ, HUANGZ, et al. Anchor3DLane: Learning to regress 3D anchors for monocular 3D lane detection[C]//2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023: 17451-17460.
[24]
GARNETTN, COHENR, PE’ERT, et al. 3D-LaneNet: End-to-end 3D multiple lane detection[C]//2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019: 2921-2930.
FANYing, SHILei, SUWeiwei, et al. Lane detection algorithm based on PINet+RESA network[J]. Journal of Jiangsu University (Natural Science Edition), 2023, 44(4): 373-378. (in Chinese)
ZHANGYing, ZHANGLulu, SUNYue, et al. Improvement of LSTR algorithm and its application in lane line detection[J]. Journal of Chinese Computer Systems, 2024, 45(8): 1863-1868. (in Chinese)
HONGShuying, ZHANGDonglin. Survey on lane line detection techniques for classifying semantic information processing modalities[J]. Computer Engineering and Applications, 2025, 61(5): 1-17. (in Chinese)
WUKaijun, XUZehao, SHANHongquan. Rapid detection method for self-exploding defects in glass insulators based on improved FasterNet and YOLOv5[J]. High Voltage Engineering, 2024, 50(5): 1865-1876. (in Chinese)
LIJuncheng, XUZengbing, SUNMaoji. Mushroom maturity detection model based on improved YOLOv5[J]. Agricultural Equipment & Vehicle Engineering, 2024, 62(6): 18-22. (in Chinese)
[35]
LIUC, ZHANGM, YANH, et al. CSYOLO: A YOLOv8-based PCB defect detection model integrating composite backbone networks and dynamic snake convolution[J]. Journal of Measurement Science and Instrumentation, 2026, 17(1): 151-161.
LIHaoxuan, SUYanqiong. Road defect detection based on improved YOLOv8[J]. Journal of Test and Measurement Technology, 2024, 38(5): 506-512. (in Chinese)