面向交通标志检测的YOLOv8n轻量化协同改进模型

doi:10.3778/j.issn.1002-8331.2507-0109

摘要/Abstract

摘要： 为应对交通标志检测中小目标易漏检、背景干扰强烈以及模型结构臃肿带来的部署障碍，提出一种基于YOLOv8n的轻量化改进检测模型RACP-YOLO。该模型引入轻量型主干C2f-RVB模块，以优化低层语义表达；采用ADown模块进行多尺度下采样，有效平衡分辨率与感受野，提升目标感知能力；结合CAA注意力机制增强通道间依赖性与目标显著性响应。在此基础上，模型在检测头部分引入SCConv结构作为核心改进，其中包含SRU（空间重建单元）与CRU（通道重建单元）双分支结构，配合新增P2分支设计成SCHead，用于增强小目标与局部空间建模能力。实验结果显示，RACP-YOLO在TT100K数据集上mAP0.5达到0.685，较YOLOv8n提升了2.1%；参数量由3.01×106降至1.12×106，压缩幅度达62.8%，计算量由8.1×109降至4.3×109，减少46.9%。在CCSTB数据集的泛化实验进一步验证了该模型在夜晚、强光及雨天等复杂场景下的适应性与检测稳定性；在检测精度提升的同时，有效降低了参数规模与计算开销，适用于车载与边缘场景的高效部署需求。

关键词: 交通标志检测, 轻量化模型, SCHead, YOLOv8改进, 小目标检测

Abstract: A lightweight object detection model, RACP-YOLO (reconstruction-aware compressed prediction YOLO), is proposed to address challenges in traffic sign detection, including missed detection of small targets, interference from complex backgrounds, and excessive model complexity. The backbone integrates a compact C2F-RVB module to improve low-level semantic representation and employs an ADown module for multi-scale downsampling, effectively balancing resolution and receptive field to enhance object perception. A channel?aware attention (CAA) mechanism is used to strengthen inter-channel dependencies and saliency response.The core improvement lies in the proposed SCConv detection head, composed of a spatial reconstruction unit (SRU) and channel reconstruction unit (CRU) in a dual-branch design. Combined with an additional P2 branch, the resulting SCHead enhances spatial modeling for small-scale and local targets. Experimental results on the TT100K dataset demonstrate that RACP-YOLO achieves a mAP0.5 of 0.685, surpassing YOLOv8n by 2.1%. The number of parameters is reduced from 3.01×106 to 1.12×106 (a reduction of 62.8%), and computational cost drops from 8.1×109 to 4.3×109 (a reduction of approximately 46.9%). Furthermore, generalization experiments on the CCSTB dataset confirm that the proposed model maintains stable detection performance and strong adaptability in complex scenarios, such as nighttime, strong light, and rainy conditions. This improvement enables higher detection accuracy while significantly enhancing model compactness and deployment efficiency, making it well-suited for real-time applications in in-vehicle and edge scenarios.

Key words: traffic sign detection, lightweight model, SCHead, YOLOv8 enhancement, small object detection

方天睿, 程光, 柳海林, 唐少虎. 面向交通标志检测的YOLOv8n轻量化协同改进模型[J]. 计算机工程与应用, 2025, 61(23): 233-247.

FANG Tianrui, CHENG Guang, LIU Hailin, TANG Shaohu. Lightweight and Synergistically Enhanced YOLOv8n Model for Traffic Sign Detection[J]. Computer Engineering and Applications, 2025, 61(23): 233-247.

参考文献

[1] 伍晓晖, 田启川. 交通标志识别方法综述[J]. 计算机工程与应用, 2020, 56(10): 20-26.
WU X H, TIAN Q C. Survey of traffic sign recognition[J]. Computer Engineering and Applications, 2020, 56(10): 20-26.
[2] SOUANI C, FAIEDH H, BESBES K. Efficient algorithm for automatic road sign recognition and its hardware implementation[J]. Journal of Real-Time Image Processing, 2014, 9(1): 79-93.
[3] BARODI A, BAJIT A, ZEMMOURI A, et al. Improved deep learning performance for real-time traffic sign detection and recognition applicable to intelligent transportation systems[J]. International Journal of Advanced Computer Science and Applications, 2022, 13(5).
[4] FLORES-CALERO M, ASTUDILLO C A, GUEVARA D, et al. Traffic sign detection and recognition using YOLO object detection algorithm: a systematic review[J]. Mathematics, 2024, 12(2): 297.
[5] HORAK K, CIP P, DAVIDEK D. Automatic traffic sign detection and recognition using colour segmentation and shape identification[J]. MATEC Web of Conferences, 2016, 68: 17002.
[6] FLEYEH H. A novel fuzzy approach for shape determination of traffic signs[C]//Proceedings of the Indian International Conference on Artificial Intelligence (IICAI), 2005: 1847-1862.
[7] TIMOFTE R, ZIMMERMANN K, VAN GOOL L. Multi-view traffic sign detection, recognition, and 3D localisation[J]. Machine Vision and Applications, 2014, 25(3): 633-647.
[8] ELLAHYANI A, MOHAMED E L, EL I, et al. Traffic sign detection and recognition using features combination and random forests[J]. International Journal of Advanced Computer Science and Applications, 2016, 7(1): 686-693.
[9] SONG Y C, GUO S Q. Traffic sign recognition based on HOG feature extraction[J]. Journal of Measurements in Engineering, 2021, 9(3): 142-155.
[10] GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2014: 580-587.
[11] GIRSHICK R. Fast R-CNN[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision. Piscataway: IEEE, 2016: 1440-1448.
[12] REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149.
[13] LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot MultiBox detector[C]//Proceedings of the 14th European Conference on Computer Vision. Cham: Springer, 2016: 21-37.
[14] LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[C]//Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway: IEEE, 2017: 2999-3007.
[15] LAW H, DENG J. CornerNet: detecting objects as paired keypoints[C]//Proceedings of the European Conference on Computer Vision. Cham: Springer, 2018: 765-781.
[16] DU C J, SU S Y, LIN C W, et al. A lightweight network for traffic sign detection via multiple scale context awareness and semantic information guidance[J]. Scientific Reports, 2025, 15: 10110.
[17] DU Q L, WU Y M, TIAN L F, et al. A lightweight traffic sign detection algorithm based on improved YOLOv7[C]//Proceedings of the 2023 4th International Conference on Intelligent Computing and Human-Computer Interaction. Piscataway: IEEE, 2023: 428-431.
[18] CHEN B X, FAN X W. MSGC-YOLO: an improved lightweight traffic sign detection model under snow conditions[J]. Mathematics, 2024, 12(10): 1539.
[19] CHENG T H, SONG L, GE Y X, et al. YOLO-world: real-time open-vocabulary object detection[C]//Proceedings of the 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2024: 16901-16911.
[20] QIU J, ZHANG W B, XU S Y, et al. DP-YOLO: a lightweight traffic sign detection model for small object detection[J]. Digital Signal Processing, 2025, 165: 105311.
[21] LI J, DENG Q W, GAO W X, et al. DSF-YOLO for robust multiscale traffic sign detection under adverse weather conditions[J]. Scientific Reports, 2025, 15: 24550.
[22] FANG S M, CHEN C M, LI Z J, et al. YOLO-ADual: a lightweight traffic sign detection model for a mobile driving system[J]. World Electric Vehicle Journal, 2024, 15(7): 323.
[23] LIN Z Y, WU Y F, MA Y H, et al. YOLO-LLTS: real-time low-light traffic sign detection via prior-guided enhancement and multi-branch feature interaction[J]. arXiv:2503.13883, 2025.
[24] WANG A, CHEN H, LIN Z J, et al. RepViT: revisiting mobile CNN from ViT perspective[J]. arXiv:2307.09283, 2023.
[25] WANG C Y, YEH I H, MARK LIAO H Y. YOLOv9: learning what you want to learn using programmable gradient information[C]//Proceedings of theEuropean Conference on Computer Vision. Cham: Springer, 2025: 1-21.
[26] CAI X H, LAI Q X, WANG Y W, et al. Poly kernel inception network for remote sensing detection[J]. arXiv:2403.06258, 2024.
[27] LI J F, WEN Y, HE L H. SCConv: spatial and channel reconstruction convolution for feature redundancy[C]//Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway: IEEE, 2023: 6153-6162.
[28] VARGHESE R, M S. YOLOv8: a novel object detection algorithm with enhanced performance and robustness[C]//Proceedings of the 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems. Piscataway: IEEE, 2024: 1-6.
[29] MA G, XU C, XU Z, et al. An improved small target detection algorithm based on YOLOv8s[J]. Radioengineering, 2025, 34(2): 206-223.