摘要
在迅速发展的现代制造业中,机器零件尺寸检测是非常重要的环节,针对目前自动焊接机焊接脚检测普遍以工人检测为主、人眼易出现视觉疲劳导致检测效率底下、错误率高等问题,提出一种基于机器视觉的五金件尺寸自动检测方法。通过图像预处理方法使得采集到的图像尽可能还原与实物相符合,使检测结果更为精确;通过有关于模板匹配的算法研究,选定基于形状的模板匹配的算法来获取目标图像的位置信息,通过抓圆、卡尺、几何测量、直线测量算法研究,来获取五金件的3个圆之间的圆心距、角度和间隙并将结果显示。实验数据表明,检测平均耗时为35 ms,圆的绝对误差平均值在0.064 cm,卡尺绝对误差平均值在0.017 cm,精度等平均在99%以上.实验结果表明,运用LabVIEW的编程原理及思想编写出一整套五金件尺寸检测系统符合自动检测的效率和精度要求。
In the rapidly developing modern manufacturing industry,the size detection of machine parts is a very important link.In view of the problem that the automatic welding machine welding feet detection is generally based on worker detection,human eyes are prone to visual fatigue resulting in lower detection efficiency and higher error rate,An automatic dimension detection method for hardware parts based on machine vision is proposed..In this paper,image preprocessing method is used to make the collected image as close as possible to the real object,so as to make the detection result more accurate;through the research of template matching algorithm,the shape-based template matching algorithm is selected to obtain the position information of the target image,and through grasping circles,calipers,geometric measurement and straightening.Line measurement algorithm is studied to obtain the center distance,angle and clearance between the three circles of hardware and the results are given.The experimental data show that the average detection time is 35 ms,the average absolute error of the circle is 0.064 cm,the average absolute error of the caliper is 0.017 cm,and the average accuracy is over 99%.The experimental results show that the programming principle and idea of LabVIEW are used to compile a set of hardware dimension detection system which accords with the efficiency and accuracy of automatic detection.
作者
谢珩
汪成龙
黄振雄
邓靖康
张文辉
王顺喜
Xie Heng;Wang Chenglong;Huang Zhenxiong;Deng Jingkang;Zhang Wenhui;Wang Shunxi(Institute of Electronic Information and electrical engineering,Huizhou University,Huizhou 516015,China)
出处
《电子测量技术》
2019年第24期152-156,共5页
Electronic Measurement Technology
基金
惠州市科技计划项目(2015B00002008)
广东省普通高校青年创新人才项目(2015KQNCX156)资助。
关键词
机器视觉
模板匹配
尺寸检测
machine vision
template matching
size detection
