基于TSD-YOLO的交通标志检测
CSTR:
作者:
作者单位:

(1. 华北理工大学 电气工程学院, 河北 唐山 063210;2. 天津城建大学 控制与机械工程学院, 天津 300384;3. 天津理工大学 电气工程与自动化学院, 天津 300384)

作者简介:

井方科(1999-),男,山东省东平县人,硕士生,主要从事智能感知与控制技术研究;

通讯作者:

中图分类号:

TP391.4

基金项目:

国家自然科学基金项目(62103298).通信作者:任红格


Traffic Sign Detection Based on TSD-YOLO
Author:
Affiliation:

(1. School of Electrical Engineering, North China University of Science and Technology, Tangshan 063210, CHN;2. School of Control and Mechanical Engineering, Tianjin Chengjian University, Tianjin 300384, CHN;3. School of Electrical Engineering and Automation, Tianjin University of Technology, Tianjin 300384, CHN)

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    针对现有交通标志检测算法对小目标检测效果较差的问题,提出了一种基于级联多尺度特征融合的交通标志检测算法。首先,设计了一种新的级联多尺度特征融合网络,利用多尺度序列特征融合模块和三重特征编码模块,使模型能更好地融合交通标志的全局特征和细节特征。其次,在骨干网络中加入可变形注意力机制,使模型专注于相关区域并捕获更丰富的图像特征。最后,使用Inner-IoU损失函数,提升了模型的泛化性能。在CCTSDB数据集上的测试结果表明,改进模型的平均精度为55.3%,较YOLOv8s模型提升了3.2%;在TT100K和VOC数据集上的表现凸显了模型出色的泛化性能。

    Abstract:

    A multiscale feature fusion-based traffic sign detection (TSD) algorithm is proposed to address the problem of poor performance of existing object detection algorithms in identifying small target traffic signs. First, a novel cascaded multiscale feature fusion network was designed, which fully utilizes the multiscale sequence feature fusion structure and triple feature encoding module, enabling the network to better integrate the detailed and global features of traffic signs. Second, deformable attention mechanisms were incorporated into the backbone network to enable the model to focus on relevant regions and capture richer image features. Finally, the use of the inner intersection over union (IoU) loss function improved the generalization performance of the TSD-YOLO model. The experimental results on the CSUST Chinese traffic sign detection benchmark (CCTSDB) dataset show that the average accuracy of the improved model was 55.3%, which is 3.2% higher than that of YOLOv8s. In addition, the performance on the visual object classes (VOC) dataset and benchmark Tsinghua-Tencent 100K (TT100K) dataset highlights the excellent generalization performance of the model.

    参考文献
    相似文献
    引证文献
引用本文

井方科,任红格,李松,史涛.基于TSD-YOLO的交通标志检测[J].半导体光电,2024,45(6):1014-1023. JING Fangke, REN Hongge, LI Song, SHI Tao. Traffic Sign Detection Based on TSD-YOLO[J].,2024,45(6):1014-1023.

复制
分享
相关视频

文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2024-07-02
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2025-02-20
  • 出版日期:
文章二维码