摘要
针对钢尺端边传统人工测量效率低、精度低等问题,提出一种基于机器视觉的测量思路,并设计出一整套的自动化识别、分类系统。该系统利用相机外部触发采集图片,利用机器视觉相关软件Halcon对目标图像进行识别和一维测量,通过C#设计出上位机界面并对测量的数据进行显示以及按照实际的生产要求进行分类,最后把分类的数据通过串口传递给下位机分类机构实现分类。试验结果表明:通过Halcon里的算法包能将端边长度为24～26mm之间的钢尺测量误差控制在0.1mm以内,并能实现快速分类。
In response to the problems of low efficiency and low precision confronting traditional manual measurement for steel ruler end face,a machine vision-based measurement idea was proposed,with a complete set of automatic identification and classification system designed.The system employed the external trigger of the camera to capture images,applied halcon,a machine vision related software,to identify and measure the target image,and designed the host computer interface through C#and displayed and classified the measured data according to the actual production requirements.Ultimately,the classified data were transmitted to the slave computer classification organization through the serial port to realize classification.The experimental results showed that the measurement error of the steel ruler with the length between 24 mm and 26 mm could be controlled within 0.1 mm through the algorithm package in halcon,and the system could realize rapid classification.
作者
肖金辉
项新建
XIAO Jinhui;XIANG Xinjian(School of Automation and Electrical Engineering,Zhejiang University of Science and Technology,Hangzhou 310023,Zhejiang,China)
出处
《浙江科技学院学报》
CAS
2018年第6期488-495,共8页
Journal of Zhejiang University of Science and Technology
基金
浙江科技学院安吉县政府对接合作项目(2015002)
关键词
机器视觉
HALCON
一维测量
分类
C#
下位机
machine vision
halcon
one dimensional measurement
classification
C#
slavecomputer
