基于机器视觉的齿轮产品外观缺陷检测

被引:19
作者
杨淑莹 [1 ]
任翠池 [2 ]
张成 [2 ]
何丕廉 [1 ]
机构
[1] 天津大学计算机科学与技术学院
[2] 天津理工大学计算机科学与技术学院
关键词
视觉检测; 图像配准; 图像融合;
D O I
暂无
中图分类号
TP274.4 [];
学科分类号
摘要
提出应用机器视觉技术对齿轮外观缺陷进行检测.首先对齿轮产品实物图像进行预处理,提取齿轮边缘图像,对其实物图像和设计图像进行分割和缩放处理,得到相同大小的实物;其次,采用互信息方法对两幅图像进行配准,通过不断旋转实物图像,计算它与设计图像的互信息,找出实物图像的最佳旋转配准角度;在此基础上将两幅图像进行小波融合,即自动对齐叠加在一起,根据重合效果,检测人员可以判断生产出的齿轮产品是否合格.通过实例验证了该方法的可行性和有效性,检测效果直观.
引用
收藏
页码:1111 / 1114
页数:4
相关论文
共 8 条
  • [1] Using local structure for the reliable removal of noise from the output of the LoG edge detector[J.]. Masoud A A,Bayoumi M M. IEEE Transactions on Systems Man and Cybernetics . 1995
  • [2] Registration of image cubes using multivariate mutual information. Kern J P,Pattichis M,Stearns S D. The Thirty-Seventh Asilomar Conference on Signals,Systems and Computers . 2003
  • [3] Data fusion of multiple polarimetric SAR images using discrete wavelet transform(DWT). Sahyun Hong,Moon W M,Hong-Yul Paik. IEEE Geoscience and Remote Sensing Letters . 2002
  • [4] Medical image registration using mutual information. Frederik Maes,Dirk Vandermeulen,Paul Suetens. Proceedings of IEEE . 2003
  • [5] Use of the Laplacian of Gaussian operator in prostate ultrasound image processing. Zhao F W,de Silva C J S. Proceedings of the 20th Annual International Conference of the IEEE . 1998
  • [6] A stable region-based multiscale image fusion scheme with thermal and visible image application for mis-registration problem. Charoentam O,Patanavijit V,Jitapunkul S. IEEE North-East Workshop on Circuits and Systems . 2006
  • [7] Multimodality image registration by maximization of mutual information[J.]. Maes F,Collignon A,Vandermeulen D. IEEE Transactions on Medical Imaging . 1997
  • [8] Mutual-information-based registration of medical images:A survey[J.]. Pluim J P W,Maintz J B A,Viergever M A. IEEE Transactions on Medical Imaging . 2003