基于YOLOv5的RFID电子标签检测研究  

Research on RFID electronic label detection based on YOLOv5

作　　者：李成海 曹梦妍 Li Chenghai;Cao Mengyan(School of Management Science and Engineering,Anhui University of Technology,Ma’anshan 243000,China;School of Business,Anhui University of Technology,Ma’anshan 243000,China)

机构地区：[1]安徽工业大学管理科学与工程学院,马鞍山243000 [2]安徽工业大学商学院,马鞍山243000

出　　处：《现代计算机》2024年第12期13-18,共6页Modern Computer

摘　　要：在无线射频识别(RFID)电子标签的生产过程中,由于制造技术的不完善、操作失误和设备问题,生产的RFID电子标签常随机出现包括墨点、胶带、空贴等多种缺陷。为了帮助工厂降本增效,加快工厂自动化生产线智能化升级,使用深度学习算法YOLOv5,结合视觉检测设备在生产线上自动检测RFID电子标签的缺陷。研究提出的RFID电子标签检测方法的测试结果显示,YOLOv5模型在缺陷检测上的准确率高达98.9%。将模型部署到生产线上,该模型展示了全覆盖、高速度和高精确度的检测性能,大大提高了RFID电子标签的检测效率。In the production process of Radio Frequency Identification(RFID)electronic labels,various defects such as ink dots,adhesive tape,and misalignment often occur randomly due to imperfect manufacturing techniques,operational errors,and equipment issues.To assist factories in reducing costs,improving efficiency,and accelerating the intelligent upgrade of automated production lines,this study employs the YOLOv5 deep learning algorithm in conjunction with visual detection devices to automatically identify defects in RFID electronic labels on the production line.The test results of the proposed RFID electronic label detection method using the YOLOv5 model show an impressive accuracy of 98.9%in defect detection.Deploying the model on the production line demonstrates comprehensive coverage,high speed,and high accuracy in defect detection,significantly enhancing the efficiency of RFID electronic label inspection.

关 键 词：RFID电子标签 YOLOv5 深度学习 缺陷检测 

分 类 号：TP391.44[自动化与计算机技术—计算机应用技术]