Application of Detecting Part's Size Online Based on Machine Vision

详细信息	查看全文 | 推荐本文 |

• 作者：Tian Yuan-yuan¹ ; Liu Si-yang¹ ; Tan Qing-chang²
• 关键词：machine vision ; super resolution reconstruction ; sub-pixel ; edge detection
• 刊名：Energy Procedia
• 出版年：2012
• 期刊代码：P136_18766102
• 类别：pw
• 出版时间：2012
• 卷：16
• 期：Complete
• 页码：1948-1956
• 文件大小：472 K

摘要

In order to use a method untouched to measure part's size, detection system is built. The research is on characteristic point detection, CCD camera calibration and distance detection between two special points. Firstly, the theory of edge detection based on Gauss point spread function is introduced. Secondly, it is calibrated using the characteristic points on the edge. Then a simple CCD calibration method of resolving the solution is deduced to solve the related parameters of model. Finally, 3D special coordinates of characteristic points detected can be obtained by the model. It is realized to detecting part size process online by calculating the distance between two special points. The experimental result shows that when physical length is 0.6 times than focal length, the given precise of measuring result can reach nearly 0.006 mm. The precise of image is stable.