| 注册
首页|期刊导航|南京信息工程大学学报|基于坐标注意力机制的轻量级安全帽佩戴检测

基于坐标注意力机制的轻量级安全帽佩戴检测

盖勇刚

南京信息工程大学学报2025,Vol.17Issue(3):315-327,13.
南京信息工程大学学报2025,Vol.17Issue(3):315-327,13.DOI:10.13878/j.cnki.jnuist.20240619001

基于坐标注意力机制的轻量级安全帽佩戴检测

Lightweight safety helmet wearing detection based on coordinate attention

盖勇刚1

作者信息

  • 1. 沈阳理工大学 自动化与电气工程学院,沈阳,110159
  • 折叠

摘要

Abstract

Existing helmet wearing detection methods exhibit poor performance in terms of accuracy and real-time capability due to problems such as dense targets and occlusion.Here,a lightweight detection model named CA-YOLO is proposed,which is designed to improve detection accuracy and real-time performance.First,the backbone network of YOLOv8 is improved using the MobileNetv3 network to reduce the number of parameters and computa-tions,thereby increasing the detection speed of the network.Then a DCNv3 module is introduced in the Neck part to enhance the model's efficiency in extracting spatial features.Second,a multi-scale feature extraction module and a Coordinate Attention(CA)module are incorporated into the YOLOv8 network.By adding global information,the feature information is enriched,thus improving the network's feature extraction capability.Finally,the Complete In-tersection over Union(CIoU)loss is replaced with the Alpha-IoU function,which accelerates the learning process of targets by setting weight coefficients to further improve the detection accuracy.Experimental results show that,com-pared to the YOLOv8 model and other existing classic and novel algorithms,the CA-YOLO model achieves an aver-age detection accuracy of 91.33%,an improvement of 0.54 percentage points over the YOLOv8 model.Additionally,the model size and the number of parameters are reduced by 41%and 39%,respectively,and the de-tection speed is increased by 16.9%.Compared to other models,the CA-YOLO model achieves a good balance be-tween accuracy and real-time performance,satisfying the requirements for detecting coal miners'safety helmet wear-ing.

关键词

目标检测/安全帽佩戴检测/YOLOv8/坐标注意力机制/轻量化

Key words

object detection/safety helmet wearing detection/YOLOv8/coordinate attention/lightweight

分类

计算机与自动化

引用本文复制引用

盖勇刚..基于坐标注意力机制的轻量级安全帽佩戴检测[J].南京信息工程大学学报,2025,17(3):315-327,13.

基金项目

辽宁省教育厅基本科研重点攻关项目(JYTZD2023006) (JYTZD2023006)

南京信息工程大学学报

OA北大核心

1674-7070

访问量0
|
下载量0
段落导航相关论文