首页 | 本学科首页   官方微博 | 高级检索  
     

基于机器视觉的装配模式识别
引用本文:覃仁超.基于机器视觉的装配模式识别[J].可编程控制器与工厂自动化(PLC FA),2006(1):105-108.
作者姓名:覃仁超
作者单位:重庆大学自动化学院
摘    要:复杂工件最主要的装配缺陷是漏装。随着现代生产技术的发展,需要检测的工件越来越复杂,需要判别的CT图像的数量也越来越大。依靠人工进行评判的方法已很难适应现代化生产的需要。文中提出一种装配缺陷的自动识别方法。首先,用相位相关法实现工件相同断面CT图像的配准,然后采用修正的Hausdorff距离判断待检测图像中可能存在模板图像的位置。最后利用模板图像在原始图像中的位置信息进一步判断装配是否正确。

关 键 词:缺陷识别  相位相关  Hausdorff距离  CT图像
文章编号:1606-5123(2006)01-0105-04

Pattern Recognition of Assemblage Based on Machine Vision
Qin Renchao.Pattern Recognition of Assemblage Based on Machine Vision[J].Programmable controller & Factory Automation(PLC & FA),2006(1):105-108.
Authors:Qin Renchao
Affiliation:Qin Renchao
Abstract:The main assembly defect of complex workpiece is lack of some parts. With the development of manufacture technique, workpieces need to be inspected become more and more complex, the number of CT image increased. The inspect method depend on manual is not suitable for modern manufacture. The paper presents an automatic defect detection method. First, registration the same position CT image with phase correlation, then locate the model image using Hausdorff distance, finally, using the position information which the model image located in the original image, to judge the assembly is right or not.
Keywords:Defect detection  Phase correlation  HausdorffdiStance  CT image
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号