This paper discusses the application of a back-propagation multi-layer perceptron and a learning vector quantization network to the classification of defects in valve stem seals for car engines.
Both networks were trained with vectors containing descriptive attributes of known flaws. These attribute vectors (‘signatures’) were extracted from images of the seals captured by an industrial vision system. The paper describes the hardware and techniques used and the results obtained.
Get full access to this article
View all access options for this article.
References
1.
WerbosP. J.Beyond regression: new tools for prediction and analysis in the behavioural sciences. PhD thesis, Harvard University, Cambridge, Mass., 1974.
2.
RumelhartD. E.HintonG. E.WilliamsR. J.Learning internal representation by error propagation. In Parallel distributed processing: explorations in the microstructures of cognition (Eds RumelhartD. E.McClellandJ. L.), Vol. I, 1986, pp. 318–362 (MIT Press, Cambridge, Mass.).
3.
KohonenT.Self-organization and associative memory, 1984 (Springer Verlag, Heildelberg).
4.
KohonenT.An introduction to neural computing. Neural Networks, 1988, 1, 3–16.
5.
TenorioM. F.HughesC. S.Real time noisy image segmentation using an artificial neural network model. Proceedings of the IEEE First International Conference on Neural networks, San Diego, Calif., 1987, Vol. 4, pp. 357–363.
6.
EichmannG.JankowskiM.StojancicM.Shape description with an associative memory. SPIE Hybrid Image Processing, 1986, 638, 76–82.
7.
YangH.GuestC. C.Performance of backpropagation for rotation invariant pattern recognition. Proceedings of the IEEE First International Conference on Neural networks, San Diego, Calif., 1987, Vol. 4, pp. 365–370.
8.
PhamD. T.Bayro-CorrochanoE. J.Neural computing for noise filtering, edge detection, and signature extraction. In Journal systems engineering, Vol. 2, 1992, pp. 111–122 (Springer Verlag, London).
9.
LansnerA.EkebergO.Reliability and speed of recall in an associative network. IEEE Trans. Pattern Anal. Machine Intell., July 1985, PAMI-7(4), 490–498.
10.
CarpenterG. A.Neural network models for pattern recognition and associative memory. Neural networks, 1989, 2(4), 243–257.
11.
AleksanderI.ThomasW. V.BowdenP. A.WISARD—a radical step forward in image recognition. Sensor Review, July 1984, pp. 120–124.
12.
GloverD. E.A hybrid optical Fourier/electronic neurocomputer machine vision inspection system. ROBOTS 12 in VISION'88 Conference Proceedings, Detroit, Mich., 5–9 June 1988, SME, Vol. 2, pp. 8–77–8–103.
DudaR. O.HartP. E.Pattern classification and scene analysis, 1973 (John Wiley, New York).
15.
GouinP. R.Performing a part inspection task with a two-staged neural network learning system. Electronic Imaging 88, International Exposition and Conference on Advance printing of paper-summaries, Boston, Mass., October 1988, Vol. 1, pp. 60–65.
16.
BeckH.McDonaldD.BrzakovicD.A self-training visual inspection system with a neural network classifier. IEEE International Conference on Neural networks, Washington, 1989, Vol. 1, pp. I-307–I-311.
17.
McCullochW. S.PittsW.A logical calculus of the ideas imminent in nervous activity. Bull. Math. Biophys., 1943, 9, 127–147.
18.
Performance Imaging Inc. Image analysis library manual, Revision 2.1, January 1992 (Performance Vision Limited, Solihull, West Midlands).
