计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (14): 12-14.DOI: 10.3778/j.issn.1002-8331.2009.14.004

• 博士论坛 • 上一篇    下一篇

机器视觉智能机器人无损检测系统

陈志国,须文波   

  1. 江南大学 信息工程学院,江苏 无锡 214122
  • 收稿日期:2008-12-08 修回日期:2009-02-16 出版日期:2009-05-11 发布日期:2009-05-11
  • 通讯作者: 陈志国

Nondestructive flaw detection system of intelligent robot with machine vision

CHEN Zhi-guo,XU Wen-bo   

  1. School of Information Engineering,Jiangnan University,Wuxi,Jiangsu 214122,China
  • Received:2008-12-08 Revised:2009-02-16 Online:2009-05-11 Published:2009-05-11
  • Contact: CHEN Zhi-guo

摘要: 针对目前无损检测过程中的大量人工操作、检测效率低下和检测精度低的特点,提出了一种搭载于工业机器人的无损检测系统的设计方法,实现了系统的软硬件设计。设计的机器人系统具有4个自由度,可以实现三维空间的立体检测,并采用自适应模糊推理系统提高检测适应性。系统采用彩色图像分割算法和自适应区域生长图像分割算法都能有效提取出缺陷区域。应用结果表明系统稳定性好、检测精度高,大大提高了检测效率。

关键词: 机器视觉, 机器人, 无损检测

Abstract: In order to solve the low efficiency and poor precision problems of traditional manual operation on nondestructive flaw detection,an integrated nondestructive flaw detection system installed on an industrial robot is described.The developed industrial robot with 4-dof can perform detection in three-dimensional space and adaptive-network-based fuzzy inference system is adopted to improve adaptability.A color image segmentation algorithm and an adaptive region growing algorithm are adopted and proved to be effective in defect region extraction.Actual results indicate that the developed system has advantages of good stability and high precision and great improvement of detection efficiency.

Key words: machine vision, robot, nondestructive flaw detection