基于机器视觉的芯片检测系统的设计与实现  

Design and Implementation of Chip Detection System Based on Machine Vision

作　　者：孙镇江 SUN Zhenjiang(Information School,Qingdao Technical College,Qingdao Shandong 266555,China)

机构地区：[1]青岛职业技术学院信息学院,山东青岛266555

出　　处：《信息与电脑》2023年第11期15-19,共5页Information & Computer

基　　金：青岛职业技术学院重点研发专项“基于‘云大物移智’的高校能源智慧监管与服务平台建设研究”(项目编号:2020ZDYF04)。

摘　　要：为提高芯片检测的可靠性、实时性、检测效率,设计一个基于计算机视觉的芯片检测系统。首先,利用YOLOv5训练芯片识别模型,结合工业级相机实时采集芯片图像;其次,将芯片图像数据发送到检测服务器设备,判断芯片实物是否存在刮痕缺陷和引脚缺失的问题;最后,根据检测结果利用机械臂将芯片放置到不同的分区,实现检测与分拣的功能。测试结果表明,该系统的准确率高较高,能够满足工业生产的要求。To improve the reliability,real-time performance,and detection efficiency of chip detection,a chip detection system based on computer vision is designed.Firstly,use YOLOv5 to train the chip recognition model and combine it with industrial grade cameras to capture chip images in real-time.Secondly,send the chip image data to the detection server device to determine whether there are scratch defects and missing pins in the physical chip.Finally,based on the detection results,a robotic arm is used to place the chip in different partitions to achieve detection and sorting functions.The test results indicate that the accuracy of the system is high and can meet the requirements of industrial production.

关 键 词：机器视觉 芯片 刮痕缺陷 引脚缺失 

分 类 号：TN40[电子电信—微电子学与固体电子学] TP391.41[自动化与计算机技术—计算机应用技术]