欢迎访问《兵工学报》官方网站,今天是

兵工学报 ›› 2011, Vol. 32 ›› Issue (7): 872-877.

• 论文 • 上一篇    下一篇

微小尺寸零件表面缺陷光学检测方法

李晓舟1, 于化东1, 于占江1, 刘岩1, 许金凯1,2   

  1. (1.长春理工大学 机电工程学院, 吉林 长春 130022;2.中国科学院 长春光学精密机械与物理研究所, 吉林 长春 130033)
  • 收稿日期:2009-11-25 修回日期:2009-11-25 上线日期:2014-05-04
  • 通讯作者: 李晓舟1 E-mail:lixiaozhoulgdx@sian.com
  • 作者简介:李晓舟(1963—)男,副教授

Optical Inspection Method for Surface Defects of Micro-components

LI Xiao-zhou1, YU Hua-dong1, YU Zhang-jiang1, LIU Yan1, XU Jin-kai1, 2   

  1. (1.School of Mechanical and Electrical Engineering,Changchun University of Science and Technology,Changchun 130022,Jilin,China;2. Changchun Institute of Optics,Fine Mechanics and Physics,Chinese Academy of Sciences,Changchun 130033,Jilin,China)
  • Received:2009-11-25 Revised:2009-11-25 Online:2014-05-04
  • Contact: LI Xiao-zhou E-mail:lixiaozhoulgdx@sian.com

摘要: 为实现微小尺寸零件表面缺陷的快速精确检验,对微小尺寸零件表面缺陷光学检测技术进行了研究。以典型微小尺寸环形零件为研究对象,从零件的表面特征、缺陷类型及检测算法出发,建立了基于机器视觉的微小尺寸零件表面缺陷光学检测系统;通过对微小尺寸环形零件表面缺陷纹理特征研究,提出图像梯度方差加权信息熵算法抑制表面随机纹理,提取微观表面缺陷,进行微小尺寸零件缺陷的检测。实验结果表明:该检测算法能够快速有效地进行微小尺寸零件表面缺陷的识别与检测,缺陷识别率达96.5%. 实验验证了理论分析及表面缺陷检测算法的正确性,可用于微小尺寸零件表面缺陷的精确检验。为实现微小尺寸零件表面缺陷的快速精确检验,对微小尺寸零件表面缺陷光学检测技术进行了研究。以典型微小尺寸环形零件为研究对象,从零件的表面特征、缺陷类型及检测算法出发,建立了基于机器视觉的微小尺寸零件表面缺陷光学检测系统;通过对微小尺寸环形零件表面缺陷纹理特征研究,提出图像梯度方差加权信息熵算法抑制表面随机纹理,提取微观表面缺陷,进行微小尺寸零件缺陷的检测。实验结果表明:该检测算法能够快速有效地进行微小尺寸零件表面缺陷的识别与检测,缺陷识别率达96.5%. 实验验证了理论分析及表面缺陷检测算法的正确性,可用于微小尺寸零件表面缺陷的精确检验。

关键词: 信息处理技术, 微小尺寸零件, 表面缺陷检测, 图像识别, 机器视觉

Abstract: In order to realize the rapid precise detection for surface defects of micro-components, the optical detection technology was studied. Taking a typical micro ring component as object and considering its surface features and defect characteristics, an optical detection system was established on the basis of machine vision and pattern recognition. Through studies of the surface flaw texture in the component, an algorithm using weighted information entropy of image gradient variance was proposed to suppress the random surface texture, extract the flaws and detect the flaws in the micro-component. Experiment results show that the algorithm can effectively detect the flaws, and the recognition rate of the defect is about 96.5%. Also, the experiments verify its correctness, rapidity and preciseness.

Key words: information processing, micro-component, surface defect detection, image recognition, machine vision

中图分类号: