This paper discusses synergistic classification systems (SCSs) for categorizing defects in wood sheets. These systems combine several classifiers to improve the performance achievable with individula classifiers. Different types of classifier could be employed in an SCS. Because of their superior generalization ability, neural classifiers were adopted in preference to rule-based classifiers. The paper describes the tuning of the parameters of the neural classifiers and the determination of the optimum number of classifier units and the best of five strategies for combining their outputs.
JordanM. I.JacobsR. A.Hierarchical mixture of experts and the EM algorithm. Neural Computation, 1994, 6, 181–214.
6.
PhamD. T.Bayro-CorrochanoE. J.Neural computing for noise filtering, edge detection and feature extraction. J. Systems Engng, 1992, 2 (2), 111–122.
7.
PhamD. T.AlcockR. J.Automatic detection of defects on birch wood boards. Proc. Instn Mech. Engrs, Part E, Journal of Process Mechanical Engineering, 1996, 210 (E1), 45–52.
8.
ZhuD.ConnersR. W.AramanP. A.3D image segmentation and recognition in an intelligent vision system. In Intelligent Robots and Computer Vision X: Algorithms and Techniques, Proceedings of the SPIE, Vol. 1607, 1991, pp. 286–297 (SPIE, Bellingham, Washington).
9.
SchmoldtD. L.LiP.AbbottA. L.Machine Vision using artificial neural networks with local 3D neighborhoods. Comput. Electronics Agriculture, 1997, 16 (3), 255–271.
10.
AlapuranenP.WestmanT.Automatic visual inspection of wood surfaces. In Proceedings of the 11th IAPR International Conference onPattern Recognition, Vol. 1, Conference A: Computer Vision and Applications, The Hague, 1992, pp. 371–374.
11.
ErsbollB. K.ConradsenK.A strategy for grading materials using a two step classification procedure. In Industrial Applications of Inspection, Metrology and Sensing, Proceedings of the SPIE, Vol. 1821, 1992, pp. 318–329 (SPIE, Bellingham, Washington).
12.
KoivoA. J.KimC. W.Classification of surface defects on wood boards. In Proceedings of the IEEE International Conference onSystems, Man and Cybernetics, Atlanta, Georgia, USA, 1986, pp. 1431–1436 (IEEE, New York).
13.
ChoT. H.ConnersR. W.A neural network approach to machine vision systems for automated industrial inspection. In Proceedings of the International Joint Conference onNeural Networks, Seattle, Washington, USA, 1991, pp. 205–211.
14.
ChoT. H.ConnersR. W.AramanP. A.A comparison of rule-based, K-nearest neighbour and neural net classifiers for automated industrial inspection. In Proceedings of the IEEE-ACM International Conference on Developing and Managing Expert System Programs, Washington, DC, 1991, pp. 202–209.
15.
PhamD. T.PhamP.AlcockR. J.Intelligent manufacturing. In Novel Intelligent Automation and Control Systems (Ed. PfeifferJ.), Vol. I, 1998 (Papierflieger, Clausthal-Zellerfeld).
16.
PhamD. T.AlcockR. J.Selection of features for wood defect classification by a multi-layer perceptron. Technical report, Intelligent Systems Laboratory, School of Engineering, University of Wales Cardiff.
PhamD. T.AlcockR. J.Artificial intelligence based techniques for processing segmented images of wood boards. Proc. Instn Mech. Engrs, Part E, Journal of Process Mechanical Engineering, 1998, 212 (E2), 119–129.
