首页|基于改进YOLOv8的电梯内电动车识别方法研究

基于改进YOLOv8的电梯内电动车识别方法研究

扫码查看
针对电梯内电动车识别存在效率低下、精度不佳的问题,提出一种结合AUGMIX图像增强技术和改进YOLOv8模型的电动车识别方法.将变形卷积层和动态稀疏注意力机制融入YOLOv8,识别更精确和高效.实验结果表明:改进后算法模型的精确率、召回率和平均精度均值分别达到了 94.5%、93%和 82.4%,电动车识别准确率达到了 95.8%,为电梯内电动车智能识别提供了理论基础.
Research on Identification Method of Electric Vehicles in Elevators Based on Improved YOLOv8
A new electric vehicle identification method combining AUGMIX and improved YOLOv8 model is proposed to address the issues of low efficiency and poor accuracy in identifying electric vehicles in elevators.The YOLOv8 model incorporates DCNv3 and BRA to identify electric vehicles with better accuracy and efficiency.The experimental results show that the precision,recall,and mean average precision of the improved algorithm model reach 94.5%,93%,and 82.4%respectively.And the accuracy of electric vehicle identification reaches 95.8%,providing a theoretical basis for intelligent recognition of electric vehicles in elevators.

machine visionAUGMIXYOLOv8DCNv3Bi-level routing attention

路成龙、冯月贵、庆光蔚

展开 >

南京市特种设备安全监督检验研究院,江苏 南京 210002

电动车识别 AUGMIX YOLOv8 变形卷积层 动态稀疏注意力机制

国家市场监督管理总局科技计划项目江苏省市场监督管理局科技计划项目

2022MK156KJ2023039

2024

机械制造与自动化
南京机械工程学会 南京机电产业(集团)有限公司

机械制造与自动化

CSTPCD
影响因子:0.29
ISSN:1671-5276
年,卷(期):2024.53(4)
  • 4