The High-Speed Tracking of Flexible Confectionery Packets using the Hough Transform
✍ Scribed by D.C. Hodgson; H. Ekerol
- Publisher
- Elsevier Science
- Year
- 1996
- Tongue
- English
- Weight
- 454 KB
- Volume
- 2
- Category
- Article
- ISSN
- 1077-2014
No coin nor oath required. For personal study only.
✦ Synopsis
The High-Speed Tracking of Flexible Confectionery Packets using the Hough Transform ne of the ways in which snack foods such as potato crisps and peanuts are presented to the market is to display them on cards hanging from the walls of shops and pubs. Currently, the Oassembl y of packets on to these display cards is performed either totally manually or semi-automatically where the packets are introduced to the assembly machine manually. They key operation in the assembly process is the location of the leading edge of the packet, gripping this edge and pulling it through a slit in the display card. This requires the measurement of the centroid coordinates and the angle of the leading edge of the packet by means of some optical system. So, for a fully automated process the difficult part of the operation is to track, by means of a vision system, the motion of a randomly oriented flexible package as it moves along a conveyor belt and also control its position and orientation so that it is aligned with the centre line of the conveyor belt. The gripper can then grasp the leading edge successfully. It is proposed to use fast-acting power air-jets to control the position of the packet, however, only the real-time motion tracking problem is being addressed here.
A transputer-based vision system has been employed as the computing platform. A modified Boundary Chain Coding (BCC) was used to extract the edge pixels from the binarized image. The edge pixels were used to calculate the centroid co-ordinates and also to determine the orientation of the leading edge of the packet by using the Hough transform. By employing parallel processing techniques, tracking of the packet's centroid and the angle of its leading edge at 60 Hz has been achieved.