基于YOLOv5的零件识别轻量化算法
Lightweight Algorithm for Part Recognition Based on YOLOv5
刘想德 1马昊2
作者信息
- 1. 重庆邮电大学国家信息无障碍工程研发中心,重庆 400000;重庆邮电大学先进制造工程学院,重庆 400000
- 2. 重庆邮电大学先进制造工程学院,重庆 400000
- 折叠
摘要
为了解决现有的基于深度学习的零件识别模型参数量过大、检测速度慢、检测精度低的问题,以YOLOv5 模型为基础,提出了结合轻量级网络和Transformer的零件识别算法.首先,设计了一种轻量级主干特征提取网络,以减少网络的参数量和计算量,并提升推理速度;其次,将Transformer模块与C3 模块融合构成C3TR模块,以增强小目标的检测能力;最后,引入噪音净化模块,通过过滤噪音来提高零件识别模型的准确率.模型的检测平均准确率和平均召回率分别达到了 86.7%和85.5%,相较原模型分别提升了和24.2%和17.4%.实验结果表明,改进后的模型在实现模型轻量化的同时,具有更快的检测速度和更高的识别准确率.
Abstract
In order to solve the problems of excessive parameter amount,slow detection speed and low de-tection accuracy of existing deep learning-based part recognition models,this paper proposes a part recogni-tion algorithm combining a lightweight network and Transformer based on the YOLOv5 model.Firstly,a lightweight trunk feature extraction network is designed to reduce the number of parameters and computa-tion of the network and to improve the inference speed;secondly,the Transformer module is fused with the C3 module to form the C3TR module to enhance the detection of small targets;finally,the noise purifica-tion module is introduced to improve the accuracy of the part recognition model by filtering the noise.The average detection accuracy and average recall of the model reach 86.7%and 85.5%,respectively,which are improved and 24.2%and 17.4%,respectively,compared with the original model.The experimental re-sults show that the improved model has faster detection speed and higher recognition accuracy while achie-ving model lightweighting.
关键词
零件识别/模型轻量化/YOLOv5/Transformer模块/噪音净化模块Key words
part identification/model light-weighting/YOLOv5/Transformer module/noise cleaning module引用本文复制引用
出版年
2024
NETL
NSTL
万方数据