基于机器视觉的工件计数系统设计  

Design of Workpiece Counting System Based on Machine Vision

作　　者：黄德成 任德均[1] HUANG Decheng;REN Dejun(School of Mechanical Engineering,Sichuan University,Chengdu 610065,China)

机构地区：[1]四川大学机械工程学院,四川成都610065

出　　处：《现代信息科技》2023年第16期171-175,共5页Modern Information Technology

摘　　要：随着新能源车的迅速发展,各汽车制造商正不断推出新产品。在此背景下,各类工件供应是否齐全成为相关企业关注的重要问题。文章设计了一种基于机器视觉的工件计数系统。该系统具有视觉成像系统,能够对多种工件进行图像采集。同时系统还设计了一套简单易操作、功能完备的工件计数软件,满足实时统计和工件种类拓展的需求。为了稳定监测工件位置信息,还设计了一种模拟光电传感器的算法替代实物光电传感器,减少系统受干扰的可能性。最后,运用YOLOv5目标检测算法,实现了99.9%的检测精度。With the rapid development of new energy vehicles,automakers are constantly introducing new products.In this context,whether the supply of various workpieces is complete has become an important issue for relevant enterprises.In this paper,a workpiece counting system based on machine vision is designed.This system has a vision imaging system that enables to carry out image acquisition of a wide range of workpieces.At the same time,the system also designs a set of easy-to-operate and fully functional workpiece counting software to meet the needs of real-time statistics and workpiece type expansion.In order to stably monitor the position information of the workpiece,this paper also designs an algorithm that simulates the photoelectric sensor to replace the physical photoelectric sensor,so as to reduce the possibility of system interference.Finally,the YOLOv5 target detection algorithm is used to achieve a detection accuracy of 99.9%.

关 键 词：机器视觉 模拟光电门 目标检测 计数软件 

分 类 号：TP389.1[自动化与计算机技术—计算机系统结构] TP391.4[自动化与计算机技术—计算机科学与技术]