Attempts are being made to automate the process of wood sheet grading by using automated visual inspection (AVI). In AVI of wood sheets, much work has been performed on the segmentation and classification stages but relatively little on the extraction of features from segmented images for defect classification purposes. This paper concentrates on feature extraction and presents 32 features potentially suitable for characterizing wood defects. The paper describes a technique for evaluating features and discusses its application to the selection of the features helpful for accurate defect classification by a multi-layer perceptron.
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.
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.
4.
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 onDeveloping and Managing Expert System Programs, Washington, DC, 1991, pp. 202–209.
5.
SchmoldtD. L.LiP.AbbottA. L.Machine vision using artificial neural networks with local 3D neighbourhoods. Comput. Electronics Agriculture, 1997, 16 (3), 255–271.
6.
CastlemanK. R.Digital Image Processing, 1979 (Prentice-Hall, Englewood Cliffs, New Jersey).
7.
LampinenJ.SmolanderS.Self-organising feature extraction in recognition of wood surface defects and colour images. Int. J. Pattern Recognition Artificial Intell., 1996, 10 (2), 97–113.
8.
ConnersR. W.McMillinC. W.LinK.Vasquez-EspinosaR. E.Identifying and locating surface defects in wood: Part of an automated lumber processing system. IEEE Trans. Pattern Analysis Mach. Intell., 1983, 5 (6), 573–583.
9.
KoivoA. J.KimC. W.Classification of surface defects on wood boards. In Proceedings of the IEEE International Conference onSystems, Man and Cybernetics, Atlanta, Georgia, 1986, pp. 1431–1436 (IEEE, New York).
10.
SobeyP. J.Automated optical grading of timber. In Optics in Agriculture, Proceedings of the SPIE, Vol. 1379, 1990, pp. 168–179 (SPIE, Bellingham, Washington).
11.
PhamD. T.CetinerB. G.A new method for describing texture. In Proceedings of the Third International Workshop onImage and Signal Processing, Manchester, 4–7 November 1996, pp. 187–190.
12.
SobeyP. J.SempleE. C.Detection and sizing visual features in wood using tonal measures and a classification algorithm. Pattern Recognition, 1989, 22 (4), 367–380.
13.
MooreD. S.The Basic Practice of Statistics, 1995 (W.H. Freeman, New York).
ChoT. H.ConnersR. W.AramanP. A.A computer vision system for automated grading of rough hardwood lumber using a knowledge-based approach. In Proceedings of the IEEE International Conference onSystems, Man and Cybernetics, Cambridge, Massachusetts, 1990, pp. 345–350 (IEEE, New York).
16.
ConnersR. W.NgC. T.ChoT. H.McMillinC. W.Computer vision system for locating and identifying defects in hardwood lumber. In Applications of Artificial Intelligence VII, Proceedings of the SPIE, Vol. 1095, Orlando, Florida, 1989, pp. 48–63 (SPIE, Bellingham, Washington).
17.
TatariS.HaettichW.Automatic recognition of defects in wood. In Advances in Image Processing, Proceedings of the SPIE, Vol. 804, The Hague, 1987, pp. 229–236 (SPIE, Bellingham, Washington).
Department of Trade and Industry Best Practice Guidelines for Developing Neural Computing Applications, 1994 (Electronics and Engineering Division, Department of Trade and Industry, London).
20.
EstevezP. A.FernandezM.AlcockR. J.PackianatherM. S.Selection of features for the classification of wood boards. In Proceedings of the Ninth International Conference onArtificial Neural Networks (ICANN), University of Edinburgh, 7–10 September 1999 (in press).
