Abstract
A method for automatic recognition of the position and orientation of parts based on machine vision and CAD 3D model is proposed. Virtual image matching is used to confirm the position and orientation of the part. By perspective imagery,virtual images of the CAD 3D model are produced according to each pose of the part. Real image generated by CCD camera is matched with virtual images of the part. The result of matching is used to confirm the relation of the part coordinates and the machine coordinates. This method is implemented on 3D coordinate measuring machines. Experiments are given to verify the method and its precision is analyzed.
| Original language | English |
|---|---|
| Pages (from-to) | 165-170 |
| Number of pages | 6 |
| Journal | Yi Qi Yi Biao Xue Bao/Chinese Journal of Scientific Instrument |
| Volume | 24 |
| Issue number | 2 |
| Publication status | Published - 2003 |
Keywords
- Recognition of position and orientation
- Machine vision
- CAD
- Image matching
- Perspective imagery
- Virtual images