改进YOLOv5s的纱管目标检测方法  

Improved YOLOv5s Method for Yarn Tube Object Detection

作　　者：姜越夫 王青[1] 吕绪山 JIANG Yuefu;WANG Qing;LYU Xushan(School of Mechanical and Electrical Engineering,Xi’an Polytechnic University,Xi’an 710699,China)

机构地区：[1]西安工程大学机电工程学院,陕西西安710699

出　　处：《机械与电子》2024年第2期29-34,共6页Machinery & Electronics

基　　金：中国纺织工业联合会科技指导性项目(2021019)。

摘　　要：为解决传统纱管分类方法效率低下、误差较高的问题,提出一种基于改进YOLOv5s算法的纱管目标识别方法。该算法融合了坐标注意力模块(CA)和Transformer模块,提出了新的SPP模块(SPP+)替换传统的SPP模块,使用加权双向特征金字塔网络(BiFPN)思想增强特征融合,并使用WIoU损失函数替代原有的损失函数。为验证改进算法性能,制作了一套纱管数据集,并基于改进YOLOv5s算法进行了纱管检测实验。实验结果表明改进的算法具有更好的识别效果。To address the issues of low efficiency and high errors in traditional yarn tube classification methods,a yarn tube object recognition method based on an improved YOLOv5s algorithm is proposed.The enhanced YOLOv5s algorithm integrates a coordinate attention(CA)module and a Transformer module,introduces a new spatial pyramid pooling plus(SPP+)module to replace the conventional spatial pyramid pooling(SPP)module,enhances feature fusion using the weighted bidirectional feature pyramid network(BiFPN)concept,and replaces the original loss function with the wise intersection over Union(WIoU)loss function.To validate the performance of the improved algorithm,a yarn tube dataset is created,and yarn tube detection experiments are conducted based on the improved YOLOv5s algorithm.The experimental results show that the improved algorithm exhibits superior recognition capabilities.

关 键 词：深度学习 纱管检测 WIoU 

分 类 号：P391.41[天文地球—地球物理学]