作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2008, Vol. 34 ›› Issue (19): 185-186,. doi: 10.3969/j.issn.1000-3428.2008.19.062

• 人工智能及识别技术 • 上一篇    下一篇

一种基于机器视觉的圆型零件检测技术

张 宇,黄亚博,焦建彬   

  1. (中国科学院研究生院工程教育学院,北京 100049)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-10-05 发布日期:2008-10-05

Detection Technique for Circular Parts Based on Machine Vision

ZHANG Yu, HUANG Ya-bo, JIAO Jian-bin   

  1. (College of Engineering, Graduate University of Chinese Academy of Sciences, Beijing 100049)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-10-05 Published:2008-10-05

摘要: 利用机器视觉技术建立圆形零件检测系统,采用简单高效的标定方法和亚像素技术实现对圆孔和圆形零件的非接触精密测量。圆孔和圆形的直径测量过程包括图像采集、图像分析处理和结果输出。其核心算法是系统校正、图像预处理、二值化和圆检测。实验结果证明,该系统实现了在线实时检测,其精度达0.01 mm,具有良好的应用前景。

关键词: 机器视觉, 亚像素, 尺寸检测, 摄像机标定

Abstract: This paper establishes a detection system for circular parts based on machine vision technique, which uses calibration method and sub-pixel technique to implement the untouched exact measure of circular holes and circular parts. The diameter measure process of circular holes and circular parts includes image capture, image analysis and disposal, and measure results output. The image compensation, image preprocess, image binarization, and circle detect methods are used in the core algorithm of this system. Experimental results prove that it realizes online real-time detection with high accuracy in 0.01 mm and has broad prospects for the application.

Key words: machine vision, sub-pixel, dimension detection, camera calibration

中图分类号: