基于机器视觉的产品质量检测  

Product Quality Testing Based on Machine Vision

作　　者：邢佳荣 张京军[1] 高瑞贞[1] XING Jia-rong;ZHAGN Jing-jun;GAO Rui-zhen(School of Information and Electrical Engineering,Hebei University of Engineering,Handan 056038,China)

机构地区：[1]河北工程大学信息与电气工程学院

出　　处：《自动化与仪表》2019年第9期28-32,36,共6页Automation & Instrumentation

基　　金：河北省自然科学基金项目(F2017402182);河北省高等学校科学研究项目(ZD2018207)

摘　　要：图像作为人类交流和认识世界的主要媒介,随着时代的发展,对图像处理技术的需求日益增加。为了获得更好的图像处理效果,该文以螺母或钢卡头为研究对象对此进行图像处理和特征提取算法。该文采用了一种用于检查螺母或钢卡头是否合格的自动检测系统,该系统基于使用相机捕获数据并使用专门为此目的开发的计算机视觉算法研究螺母或钢卡头特征。采买多个螺母测试了该检验方法。原型系统表现良好,因为检测算法能够准确地检测特征并表征其是否合格。使用具有不同形状,尺寸和颜色的螺母进行检测。为检测性状而开发的算法证明是稳健的。As the main medium for human communication and understanding of the world,the demand for image processing technology is increasing with the development of the times. In order to obtain better image processing effect,this paper takes nut or steel chuck as the research object to carry out image processing and feature extraction algorithm. We have adopted an automatic detection system for checking whether nuts or steel chucks are qualified. The system is based on capturing data using a camera and studying the characteristics of nuts or steel chucks using a computer vision algorithm specially developed for this purpose. We bought several nuts to test the test method. The prototype system performs well because the detection algorithm can accurately detect features and characterize whether they are qualified. Use nuts with different shapes,sizes and colors for testing. The algorithm developed to detect traits has proved to be robust.

关 键 词：机器视觉 图像处理 质量检测 二值化 

分 类 号：TP271.4[自动化与计算机技术—检测技术与自动化装置]