改进 YOLOv5s的交通多目标检测方法
DOI:
CSTR:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

TP391.4

基金项目:

国家自然科学基金(62071240,62106111)项目资助


Trafficmulti-targetdetectionmethodofYOLOv5sisimproved
Author:
Affiliation:

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:

    为了提高交通目标检测的精度和效率,提出一种改进YOLOv5s的交通场景多目标检测方法,在YOLOv5s 的主干网 络中引入高效的层聚合网络结构来提高模型学习目标特征的能力,引入了通道注意力和空间注意力结合的卷积注意力模块 (BAM)机制,进一步提高网络模型的特征提取能力,通过采用a-IoU 作为边界框回归损失函数,提高了边界框回归精度。实 验结果表明,改进的目标检测模型相较于YOLOv5s原模型在检测精度上提升了2.4%,模型参数量和模型大小分别降低了 20.9%和19.1%。实现了在不同时间段准确且高效的检测交通场景的多种目标,保证了实时检测的应用需求。

    Abstract:

    In order to improve the accuracy and efficiency of trafic intersection target detection,this paper proposes an improved trafic scene target detection model based on YOLOv5s.An efficient layer aggregation network structure is introduced into the backbone network of YOLOv5s to improve the ability to learn target features.The attention mechanism CBAM of channel attention and spatial attention is introduced to further improve the feature extraction ability of the network model.The a-loU is used as the bounding box regression loss function to improve the bounding box regression accuracy.The experimental results show that compared with the original YOLOv5s model,the improved object detection model proposed in this paper has the detection accuracy increased by 2.4%and the model parameter number and model size reduced by 20.9%and 19.1%respectively.It realizes all kinds of targets of accurate and efficient detection of traffic intersection scenes in different time periods,and ensures the application requirements of real-time detection.

    参考文献
    相似文献
    引证文献
引用本文

单慧琳,吕 宗 奎,付 相 为,王 煜,张培琰,孙佳琪.改进 YOLOv5s的交通多目标检测方法[J].国外电子测量技术,2023,42(4):8-15

复制
分享
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:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2024-10-29
  • 出版日期:
文章二维码