基于深度学习的轻量化道路交通锥目标检测算法

古龙毅; 张慧; 蔡小明; 李明; 王钦若

doi:10.12052/gdutxb.240101

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于深度学习的轻量化道路交通锥目标检测算法

计算机科学与技术 | 更新时间：2026-05-27

- 基于深度学习的轻量化道路交通锥目标检测算法
- Lightweight Road Traffic Cone Detection Algorithm Based on Deep Learning
- 广东工业大学学报 2025年42卷第4期页码：39-47
- 作者机构：
  
  广东工业大学自动化学院,广东,广州,510006
- 作者简介：
  
  古龙毅(2000–) ，男，硕士研究生，主要研究方向为深度学习，E-mail：2112204062@mail2.gdut.edu.cn
  张慧(1979–) ，女，讲师，博士，主要研究方向为时滞系统优化控制、电机控制系统、柔性系统控制、深度学习，E-mail：zh2008@gdut.edu.cn
- 基金信息：
  
  广东省重点领域研发计划项目(2022B0701180001)
- DOI：10.12052/gdutxb.240101
  中图分类号： TP391
- 纸质出版：2025
- 稿件说明：
移动端阅览
古龙毅, 张慧, 蔡小明, 等. 基于深度学习的轻量化道路交通锥目标检测算法[J]. 广东工业大学学报, 2025, 42(4): 39–47.

Gu Longyi, Zhang Hui, Cai Xiaoming, et al. Lightweight Road Traffic Cone Detection Algorithm Based on Deep Learning[J]. 2025, 42(4): 39-47.
古龙毅, 张慧, 蔡小明, 等. 基于深度学习的轻量化道路交通锥目标检测算法[J]. 广东工业大学学报, 2025, 42(4): 39–47. DOI： 10.12052/gdutxb.240101

Gu Longyi, Zhang Hui, Cai Xiaoming, et al. Lightweight Road Traffic Cone Detection Algorithm Based on Deep Learning[J]. 2025, 42(4): 39-47. DOI： 10.12052/gdutxb.240101.

摘要

道路交通锥是用于高速公路和城市道路维护的安全隔离标志，采用自动化设备对其进行收放可提高作业效率，并降低传统人工收放方式带来的作业人员安全隐患。对交通锥进行精确的目标检测是确保自动化收放设备正常工作的关键。为提高识别效率，同时保证模型检测准确性并降低模型计算量和内存需求，本文提出了一种轻量化道路交通锥检测的TraCone-YOLO(TraCone-You Only Look Once) 算法。首先，采用基于共享深度可分离卷积的高效检测头，使模型在推理时更加高效；其次，在特征提取过程中使用RepC2f模块，增强模型提取多尺度特征信息的能力；最后，通过引入ELA(Efficient Local Attention) 注意力模块，提出ELA-HSFPN(ELA-Hierarchical Scale-based Feature Pyramid Network) 网络结构进行多尺度特征融合，进一步增强模型的特征表达能力。针对原有的交通锥数据集TraCon缺少不同天气和光照强度的样本，本文对TraCon交通锥数据集进行了数据增强，以模拟实际场景中的不同天气条件，有效扩充训练样本，从而提升模型的泛化能力和鲁棒性。实验结果表明，改进后的模型与原模型相比，参数量和计算量分别减少了56.2%和55.6%，均值平均精度达92.6%，与原模型相比仅下降0.4个百分点。该算法在实现模型轻量化的同时，能够保持良好的检测性能。

Abstract

Road traffic cones are safety isolation signs used for highway and urban road maintenance. Automated equipment can greatly improve operational efficiency and reduce safety hazards for workers caused by traditional manual retrieval methods. Accurate target detection of traffic cones is key to ensuring the normal operation of automated retrieval equipment. To improve recognition efficiency while ensuring model detection accuracy and reducing model computation

a lightweight road traffic cone detection algorithm called TraCone-YOLO (TraCone-You Only Look Once) is proposed in this paper. Firstly

an efficient detection head based on shared depthwise separable convolution is adopted in this paper to make the model more efficient in inference. Secondly

to improve the model's ability to extract multi-scale feature information

the RepC2f module is used in the feature extraction process. Finally

by introducing the ELA (Efficient Local Attention) attention module

an ELA-HSFPN (ELA-Hierarchical Scale based Feature Pyramid Network) network structure is proposed for multi-scale feature fusion

further enhancing the model's feature expression ability.In response to the lack of samples with different weather and light intensities in the original traffic cone dataset called TraCon

data augmentation method is applied to simulate different weather conditions in actual scenarios. The training samples are effectively expanded

thereby improving the model's generalization ability and robustness. The experimental results show that the improved model reduces parameters and computation by 56.2% and 55.6% respectively

with an average precision of 92.6%

which is only 0.4 percentage points lower than the original model. The algorithm proposed in this research can make model lightweight while maintaining good detection performance.

关键词

Keywords

references

LIU W, ANGUELOV D, ERHAN D, et al . SSD: single shot multibox detector[C ] //Computer Vision–ECCV 2016: 14th European Conference. Amsterdam: Springer International Publishing, 2016: 21-37.

ZHAI S, SHANG D, WANG S, et al . DF-SSD: an improved SSD object detection algorithm based on DenseNet and feature fusion[J ] . IEEE access, 2020, 8: 24344-24357.

CHOI H T, LEE H J , KANG H, et al . SSD-EMB: an improved SSD using enhanced feature map block for object detection[J ] . Sensors, 2021, 21(8): 2842.

REDMON J, DIVVALA S, GIRSHICK R, et al . You only look once: unified, real-time object detection[C ] //2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE, 2016: 779-788.

REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Hawaii: IEEE, 2017: 6517-6525.

LIN T Y, GOYAL P, GIRSHICK R, et al . Focal loss for dense object detection[C ] //2017 IEEE International Conference on Computer Vision (ICCV). Venice: IEEE, 2017: 2980-2988.

GIRSHICK R. Fast R-CNN[C]//2015 IEEE International Conference on Computer Vision (ICCV). Santiago: IEEE, 2015: 1440-1448.

REN S, HE K, 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, 2016, 39(6): 1137-1149.

LI Y, CHEN Y, WANG N, et al . Scale-aware trident networks for object detection[C ] //2019 IEEE/CVF International Conference on Computer Vision (ICCV). Seoul: IEEE, 2019: 6053-6062.

ZHAO Z S, QIN Y Y, LI G, et al . Traffic cone color detection based on Yolo algorithm[J ] . Automobile Applied Technology, 2022, 47(18): 28-32.

ZHAO Z S, SANG H F. Traffic cone detection system based on improved YOLOv5[J]. Journal of Electronic Measurement and Instrumentation, 2023, 37(2): 56-64.

DHALL A, DAI D, VAN GOOL L. Real-time 3D traffic cone detection for autonomous driving[C]//2019 IEEE Intelligent Vehicles Symposium (IV). Paris: IEEE, 2019: 494-501.

SU Q, WANG H, XIE M, et al . Real-time traffic cone detection for autonomous driving based on YOLOv4[J ] . IET Intelligent Transport Systems, 2022, 16(10): 1380-1390.

ALBARÁÑEZ MARTÍNEZ J, LLOPIS-IBOR L, HERNÁNDEZ-GARCÍA S, et al . A case of study on traffic cone detection for autonomous racing on a jetson platform[C ] //Iberian Conference on Pattern Recognition and Image Analysis(IbPRIA). Cham: Springer International Publishing, 2022: 629-641.

GENNA M, FAWCETT R, PRISACARIU V A. Dsconv: Efficient convolution operator[C]//2019 IEEE/CVF International Conference on Computer Vision (ICCV). Seoul: IEEE, 2019: 5148-5157.

HAN K, WANG Y, TIAN Q, et al . GhostNet: more features from cheap operations[C ] //2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE, 2020: 1580-1589.

CHEN Y, ZHANG C, CHEN B, et al . Accurate leukocyte detection based on deformable-DETR and multi-level feature fusion for aiding diagnosis of blood diseases[J ] . Computers in Biology and Medicine, 2024, 170: 107917.

KATSAMENIS I, KAROLOU E E, DAVRADOU A, et al . TraCon: a novel dataset for real-time traffic cones detection using deep learning[C ] //Novel Intelligent Digital Systems: Proceedings of the 2nd International Conference(NiDS). Cham: Springer International Publishing, 2023: 382-391.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于生成式样本合成的工件缺陷样本数据增强

基于编解码器模型的车道识别与车辆检测算法

基于数字混频的脉冲式激光测距方法的研究与实现

面向灾后救援的多无人机通感算一体化资源分配优化方法

多套餐共享储能模式下考虑功率互济的社区联盟优化运行策略