期刊文献+

基于YOLOv8的机械臂核酸咽拭子自动采集系统研究

Research on Automatic Nucleic Acid Throat Swab Collection System of Robotic Arm Based on YOLOv8
下载PDF
导出
摘要 传统核酸咽拭子采集方法是由医护人员进行人工采集,耗费大量医疗资源,且存在较高的感染风险。现有的自动机械臂核酸采样系统大多采用固定位置的一次性咬口器定位口腔,实际使用不灵活。为了降低采集成本与医护人员感染风险,设计了一种基于YOLOv8的机械臂核酸咽拭子自动采集系统。该系统采用YOLOv8目标检测算法进行咽部区域识别,结合深度相机进行咽部三维定位,当被采集者进入机械臂工作空间后,由视觉伺服系统和力控反馈系统控制机械臂夹取咽拭子完成咽部核酸咽拭子自动采集。实验结果表明,系统对咽部区域的识别率较高,力控反馈效果良好,能够满足现场使用需求。 The traditional nucleic acid throat swab collection method is manually collected by medical staff,which consumes a lot of medical resources and has a high risk of infection.Most of the existing automatic robotic arm nucleic acid sampling systems use disposable bite devices to locate the mouth,but it is not flexible in actual use.In order to reduce the collection cost and the infection risk of medical staff,a robotic arm automatic nucleic acid throat swab collection system based on YOLOv8 is designed.This system adopts the YOLOv8 object detection algorithm for identification of pharyngeal region,and combines with a depth camera for pharyngeal 3D localization.When the collector enters the working space of the robotic arm,the visual servo system and the force control feedback system control the robotic arm to clamp the throat swab to complete the automatic collection of the pharyngeal nucleic acid throat swab.The experimental results show that this system has a high recognition rate of the pharyngeal region,and the force control feedback effect is good,which can meet the needs of field use.
作者 王展 刘云啸 黎彬鸿 吴苑莎 刘浩洋 张子杰 黄宝玲 余果 WANG Zhan;LIU Yunxiao;LI Binhong;WU Yuansha;LIU Haoyang;ZHANG Zijie;HUANG Baoling;YU Guo(School of Biomedical Engineering,Guangdong Medical University,Dongguan 523808,China)
出处 《现代信息科技》 2024年第22期177-182,188,共7页 Modern Information Technology
基金 广东大学生科技创新培育专项资金资助项目(pdjh2022b0218)。
关键词 机械臂 深度相机 机器视觉 YOLOv8 力控反馈 自主定位 robotic arm depth camera Machine Vision YOLOv8 force controlled feedback autonomous positioning
  • 相关文献

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部