

浏览全部资源
扫码关注微信
广西大学 电气工程学院, 广西 南宁 530004
Received:09 February 2026,
Revised:2026-02-20,
Published:05 March 2026
移动端阅览
SONG Shaojian, LI Hao, LI Gang, et al. YOLO-DyMiF: a dynamic multi-scale traffic sign detection network for low-computing-power platforms[J]. Chinese Journal of Liquid Crystals and Displays, 2026, 41(3): 388-401.
SONG Shaojian, LI Hao, LI Gang, et al. YOLO-DyMiF: a dynamic multi-scale traffic sign detection network for low-computing-power platforms[J]. Chinese Journal of Liquid Crystals and Displays, 2026, 41(3): 388-401. DOI: 10.37188/CJLCD.2026-0021. CSTR: 32172.14.CJLCD.2026-0021.
为了解决自动驾驶场景中交通标志目标体积小、易被环境干扰而导致检测精度低,以及车载平台算力和功耗有限、难以支撑复杂模型的问题,本文提出了一种改进的轻量化检测算法YOLO-DyMiF(Dynamic Mixer and Feature Fusion)。该模型在YOLOv10n的基础上进行了两方面改进:首先,设计一种基于动态高效卷积(Adaptive Efficient Conv, AEConv)的高效动态混合器(Efficient Dynamic Mixer Structure, EDMS),并将其嵌入C3k2模块以构建C3k2_EDMS模块,用于替换YOLOv10n模型中的C2f模块,在保持主干网络特征表达能力的前提下有效压缩参数规模;其次,设计了以分层多尺度空间增强模块(Hierarchical Multi-scale Spatial Enhancement, HMSE)为核心的动态特征融合颈部网络,它通过跨层交互和自适应加权融合增强多尺度特征表征能力,在兼顾中、大目标检测性能的同时提升小目标交通标志检测精度。在TT100K数据集上的实验结果表明,与当前领先的Mamba-YOLOt相比,YOLO-DyMiF算法的mAP50提高1%,模型参数量下降了58.3%,计算量下降了42.3%。所提出的模型能够在确保高检测精度的同时显著降低计算成本,可以为自动驾驶场景中的交通标志检测提供可靠的技术支持。
To address the low detection accuracy caused by the small size of traffic signs in autonomous driving scenarios and their susceptibility to environmental interference, as well as the limited computing capability and power budget of onboard platforms that make complex models difficult to deploy, an improved lightweight detector named YOLO-DyMiF (Dynamic Mixer and Feature Fusion) is proposed. The proposed model, based on YOLOv10n, introduces two major improvements. Firstly, an Efficient Dynamic Mixer Structure (EDMS) based on Adaptive Efficient Convolution (AEConv) is designed and embedded into the C3k2 module to get a new module named C3k2_EDMS, which replaces the C2f module in YOLOv10n. This design effectively reduces the parameter scale while preserving the feature representation capability of the backbone network. Secondly, a dynamic feature fusion neck network is developed with the Hierarchical Multi-scale Spatial Enhancement (HMSE) module. Through cross-layer interactions and adaptive weighted fusion, the neck enhances multi-scale feature representation, improving the detection accuracy of small traffic signs while keeping detection performance on medium and large objects. Experimental results on the TT100K dataset show that, in comparison with the state-of-the-art Mamba-YOLOt, YOLO-DyMiF improves mAP50 by 1.0%, reduces the number of parameters by 58.3%, and decreases computational cost by 42.3%. The proposed model significantly reduces the computational cost while ensuring high detection accuracy, which provides reliable technical support for traffic sign detection in autonomous driving applications.
XU J , DU Y J , YI Y W , et al . An improved lightweight algorithm for traffic sign detection [J]. Scientific Reports , 2025 , 15 ( 1 ): 33554 . doi: 10.1038/s41598-025-18469-x http://dx.doi.org/10.1038/s41598-025-18469-x
ZHAO Z C , DU J X , LI C L , et al . Dense Tiny object detection: a scene context guided approach and a unified benchmark [J]. IEEE Transactions on Geoscience and Remote Sensing , 2024 , 62 : 5606913 . doi: 10.1109/tgrs.2024.3357706 http://dx.doi.org/10.1109/tgrs.2024.3357706
LIU Y Y , PENG J Y , XUE J H , et al . TSingNet: scale-aware and context-rich feature learning for traffic sign detection and recognition in the wild [J]. Neurocomputing , 2021 , 447 : 10 - 22 . doi: 10.1016/j.neucom.2021.03.049 http://dx.doi.org/10.1016/j.neucom.2021.03.049
WANG P Z , MOHAMED R , MUSTAPHA N , et al . YOLO-AEF: traffic sign detection on challenging traffic scenes via adaptive enhancement and fusion [J]. Neurocomputing , 2025 , 655 : 131430 .
LIANG T L , GLOSSNER J , WANG L , et al . Pruning and quantization for deep neural network acceleration: a survey [J]. Neurocomputing , 2021 , 461 : 370 - 403 . doi: 10.1016/j.neucom.2021.07.045 http://dx.doi.org/10.1016/j.neucom.2021.07.045
DING Y F , FENG W L , CHEN C Y , et al . Reg-PTQ: regression-specialized post-training quantization for fully quantized object detector [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Seattle : IEEE , 2024 : 16174 - 16184 . doi: 10.1109/cvpr52733.2024.01531 http://dx.doi.org/10.1109/cvpr52733.2024.01531
RAZA S M , ABIDI S M H , MASUDUZZAMAN M , et al . Lightweight deep learning for visual perception: a survey of models, compression strategies, and edge deployment challenges [J]. Neurocomputing , 2025 , 656 : 131357 . doi: 10.1016/j.neucom.2025.131357 http://dx.doi.org/10.1016/j.neucom.2025.131357
HAO C Y , ZHANG H , SONG W R , et al . SliNet: slicing-aided learning for small object detection [J]. IEEE Signal Processing Letters , 2024 , 31 : 790 - 794 . doi: 10.1109/lsp.2024.3373261 http://dx.doi.org/10.1109/lsp.2024.3373261
TANG J X , CHENG J D , XIANG D L , et al . Large-difference-scale target detection using a revised Bhattacharyya distance in SAR images [J]. IEEE Geoscience and Remote Sensing Letters , 2022 , 19 : 4506205 . doi: 10.1109/lgrs.2022.3161931 http://dx.doi.org/10.1109/lgrs.2022.3161931
LI Y S , LIU F X , QIN Y S , et al . Small object detection network based on progressive enhanced multi-level feature fusion [J]. Neurocomputing , 2025 , 649 ( 13 ): 130755 . doi: 10.1016/j.neucom.2025.130755 http://dx.doi.org/10.1016/j.neucom.2025.130755
ZHU Z Q , ZHENG R Z , QI G Q , et al . Small object detection method based on global multi-level perception and dynamic region aggregation [J]. IEEE Transactions on Circuits and Systems for Video Technology , 2024 , 34 ( 10 ): 10011 - 10022 . doi: 10.1109/tcsvt.2024.3402097 http://dx.doi.org/10.1109/tcsvt.2024.3402097
GAO Y X , WANG Y C , ZHANG Y X , et al . Feature super-resolution fusion with cross-scale distillation for small-object detection in optical remote sensing images [J]. IEEE Geoscience and Remote Sensing Letters , 2024 , 21 : 6008105 . doi: 10.1109/lgrs.2024.3372500 http://dx.doi.org/10.1109/lgrs.2024.3372500
CHEN J Z , HUANG H Q , ZHANG R H , et al . YOLO-TS: real-time traffic sign detection with enhanced accuracy using optimized receptive fields and anchor-free fusion [J]. IEEE Transactions on Intelligent Transportation Systems , 2025 , 26 ( 11 ): 19995 - 20011 . doi: 10.1109/tits.2025.3597710 http://dx.doi.org/10.1109/tits.2025.3597710
XU Y L , XIAO X X . PST-YOLO: a lightweight, robust approach for traffic sign detection via polynomial-based and progressive scale fusions [J]. Digital Signal Processing , 2026 , 169 : 105745 . doi: 10.1016/j.dsp.2025.105745 http://dx.doi.org/10.1016/j.dsp.2025.105745
李鹏飞 , 熊召新 , 王桂宝 . YOLOv8-NTS:一种针对交通标志检测的目标识别方法 [J]. 液晶与显示 , 2025 , 40 ( 12 ): 1868 - 1880 . doi: 10.37188/cjlcd.2025-0199 http://dx.doi.org/10.37188/cjlcd.2025-0199
LI P F , XIONG Z X , WANG G B . YOLOv8-NTS: a target detection approach for traffic sign recognition [J]. Chinese Journal of Liquid Crystals and Displays , 2025 , 40 ( 12 ): 1868 - 1880 . (in Chinese) . doi: 10.37188/cjlcd.2025-0199 http://dx.doi.org/10.37188/cjlcd.2025-0199
QU S M , YANG X Y , ZHOU H F , et al . Improved YOLOv5-based for small traffic sign detection under complex weather [J]. Scientific Reports , 2023 , 13 ( 1 ): 16219 . doi: 10.1038/s41598-023-42753-3 http://dx.doi.org/10.1038/s41598-023-42753-3
宋绍剑 , 夏海姐 , 李刚 . YOLOv5的改进算法及其在自动驾驶多目标检测的应用研究 [J]. 计算机工程与应用 , 2023 , 59 ( 15 ): 68 - 75 . doi: 10.3778/j.issn.1002-8331.2302-0185 http://dx.doi.org/10.3778/j.issn.1002-8331.2302-0185
SONG S J , XIA H J , LI G . Research on improved YOLOv5 algorithm and its application in multi-object detection for automatic driving [J]. Computer Engineering and Applications , 2023 , 59 ( 15 ): 68 - 75 . (in Chinese) . doi: 10.3778/j.issn.1002-8331.2302-0185 http://dx.doi.org/10.3778/j.issn.1002-8331.2302-0185
SUN Y , LI X , ZHAO D J , et al . Evolving traffic sign detection via multi-scale feature enhancement, reconstruction and fusion [J]. Digital Signal Processing , 2025 , 160 : 105028 . doi: 10.1016/j.dsp.2025.105028 http://dx.doi.org/10.1016/j.dsp.2025.105028
YIN C F , WANG Z H , LI Y C , et al . SBEW-YOLOV8: a small object detection algorithm for autonomous driving based on multi-scale feature fusion [J]. The Journal of Supercomputing , 2025 , 81 ( 10 ): 1125 . doi: 10.1007/s11227-025-07577-0 http://dx.doi.org/10.1007/s11227-025-07577-0
王伯霄 , 宋延嵩 , 董小娜 . 难点注意力感知红外小目标检测网络 [J]. 中国光学(中英文) , 2024 , 17 ( 3 ): 538 - 547 . doi: 10.37188/co.2023-0178 http://dx.doi.org/10.37188/co.2023-0178
WANG B X , SONG Y S , DONG X N . Indistinguishable points attention-aware network for infrared small object detection [J]. Chinese Optics , 2024 , 17 ( 3 ): 538 - 547 . (in Chinese) . doi: 10.37188/co.2023-0178 http://dx.doi.org/10.37188/co.2023-0178
YANG L H , GU Y , FENG H . Multi-scale feature fusion and feature calibration with edge information enhancement for remote sensing object detection [J]. Scientific Reports , 2025 , 15 ( 1 ): 15371 . doi: 10.1038/s41598-025-99835-7 http://dx.doi.org/10.1038/s41598-025-99835-7
TONG K , WU Y Q . Small object detection using deep feature learning and feature fusion network [J]. Engineering Applications of Artificial Intelligence , 2024 , 132 : 107931 . doi: 10.1016/j.engappai.2024.107931 http://dx.doi.org/10.1016/j.engappai.2024.107931
AHMAD I , LU W , CHEN S B , et al . Lightweight oriented object detection with dynamic smooth feature fusion network [J]. Neurocomputing , 2025 , 628 : 129725 . doi: 10.1016/j.neucom.2025.129725 http://dx.doi.org/10.1016/j.neucom.2025.129725
XU S , LI Y J , LIN M B , et al . Q-DETR: an efficient low-bit quantized detection transformer [C]. 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Vancouver, Canada : IEEE , 2023 : 3842 - 3851 . doi: 10.1109/cvpr52729.2023.00374 http://dx.doi.org/10.1109/cvpr52729.2023.00374
ZHANG S , CHE S B , LIU Z , et al . A real-time and lightweight traffic sign detection method based on ghost-YOLO [J]. Multimedia Tools and Applications , 2023 , 82 ( 17 ): 26063 - 26087 . doi: 10.1007/s11042-023-14342-z http://dx.doi.org/10.1007/s11042-023-14342-z
王龙春 , 方巍 , 张丽娟 , 等 . 基于YOLOv8s改进的自动驾驶目标检测 [J]. 液晶与显示 , 2025 , 40 ( 5 ): 773 - 784 . doi: 10.37188/cjlcd.2024-0290 http://dx.doi.org/10.37188/cjlcd.2024-0290
WANG L C , FANG W , ZHANG L J , et al . Improved autonomous driving object detection based on YOLOv8s [J]. Chinese Journal of Liquid Crystals and Displays , 2025 , 40 ( 5 ): 773 - 784 . (in Chinese) . doi: 10.37188/cjlcd.2024-0290 http://dx.doi.org/10.37188/cjlcd.2024-0290
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 . doi: 10.1016/j.dsp.2025.105311 http://dx.doi.org/10.1016/j.dsp.2025.105311
DOHERTY J , GARDINER B , KERR E , et al . BiFPN-YOLO: One-stage object detection integrating Bi-Directional Feature Pyramid Networks [J]. Pattern Recognition , 2025 , 160 : 111209 . doi: 10.1016/j.patcog.2024.111209 http://dx.doi.org/10.1016/j.patcog.2024.111209
WANG C Y , YEH I H , LIAO H Y M . YOLOv9: learning what you want to learn using programmable gradient information [C]// Proceedings of 18th European Conference on Computer Vision-ECCV 2024 . Milan, Italy : Springer , 2024 : 1 - 21 . doi: 10.1007/978-3-031-72751-1_1 http://dx.doi.org/10.1007/978-3-031-72751-1_1
ZHU Z , LIANG D , ZHANG S H , et al . Traffic-sign detection and classification in the wild [C]. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Las Vegas, USA : IEEE , 2016 : 2110 - 2118 . doi: 10.1109/cvpr.2016.232 http://dx.doi.org/10.1109/cvpr.2016.232
0
Views
400
下载量
0
CSCD
Publicity Resources
Related Articles
Related Author
Related Institution
京公网安备11010802024621