DocumentCode :
2725854
Title :
Computational Intelligence for Automated Keg Identification and Deformnation Detection
Author :
Campbell, Duncan ; Keir, Andrew ; Lees, Michael
Author_Institution :
Sch. of Eng. Syst., Queensland Univ. of Technol., Brisbane, Qld.
fYear :
2007
fDate :
1-5 April 2007
Firstpage :
77
Lastpage :
82
Abstract :
A machine vision based keg inspection system can allow cost effective keg tracking and preventative maintenance programs to be implemented, leading to substantial savings for breweries with large keg fleets. A robust keg serial number recognition and keg condition assessment process is required to cater for different keg brands and a range of keg ages in the fleet. It has been demonstrated that the proposed image processing methodology, and neural network based number recognition system, successfully located and identified keg serial numbers with a 92% digit accuracy. Furthermore, the vision system allowed the concurrent assessment of the keg condition by assessing deformity of the keg rim, and that of the filler valve. A correlation coefficient, generated using a template matching process, proved to be a suitable metric which adequately indicated rims within and outside acceptable deformity bounds
Keywords :
breweries; character recognition; computer vision; containers; image matching; inspection; neural nets; automated keg identification; breweries; computational intelligence; correlation coefficient; deformation detection; deformity bound; filler valve; image processing; keg condition assessment; keg inspection system; keg serial number recognition; keg tracking; machine vision; neural network; number recognition system; preventive maintenance; template matching; Computational intelligence; Costs; Image processing; Image recognition; Inspection; Machine vision; Neural networks; Preventive maintenance; Robustness; Valves; Keg Deformation Detection; Keg Tracking; Neural Networks; OCR;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computational Intelligence in Image and Signal Processing, 2007. CIISP 2007. IEEE Symposium on
Conference_Location :
Honolulu, HI
Print_ISBN :
1-4244-0707-9
Type :
conf
DOI :
10.1109/CIISP.2007.369297
Filename :
4221398
Link To Document :
بازگشت