摘要
结合人工智能技术和物联网技术,研究并设计了一种PCB表面缺陷检测系统,系统采用树莓派作为上位机,借助百度EasyDL定制的AI开发平台训练模型并部署在树莓派上,实现缺陷智能检测,采用PLC作为下位机控制系统机械运动,同时开发管理平台。结果表明该系统能自动检测并标识缺陷位置,存储检测数据,便于管理人员分析不良品,整个操作过程简单快捷,自动化程度高,对PCB表面缺陷检测具有一定价值。该系统还可以训练出用于其他场景的检测模型,可移植性强。
Combining artificial intelligence technology and the Internet of Things technology,a PCB surface defect detection system was studied and designed.The system uses Raspberry Pi as the upper computer,trains the model with the AI development platform customized by Baidu EasyDL,and deploys it on Raspberry Pi to achieve intelligent defect detection.A PLC is used as the lower computer to control the mechanical movement of the system,and a management platform is developed.The results show that the system can automatically detect and mark the defect position.Storing inspection data makes it easy for management personnel to analyze defective products.The entire operation process is simple,fast,and highly automated,which has certain practical value for PCB surface defect detection.The system can also train detection models for other scenarios,with strong portability.
作者
潘春玲
PAN Chuning(Quanzhou Vocational College of Economics and Business,Quanzhou,Fujian,China 362000)
出处
《湖南邮电职业技术学院学报》
2023年第3期29-33,共5页
Journal of Hunan Post and Telecommunication College
基金
2021年福建省中青年教师教育科研项目(科技类)“基于AIoT的PCB表面缺陷检测系统设计”(项目编号:JAT210937)。
