Abstract
An automatic log classification system is described. The classification was based on texture features obtained from measurements of growth rings from a single X-ray scan along each log. The speed of the log feeding restricted the possibility of full-scale scanning and image reconstruction. A preliminary analysis to study the projection of growth rings was performed by developing a simulated X-ray system. Then the projection images of 347 logs obtained in a medical scanner were used in the classification. It was not possible to measure the rings exactly due to the low resolution of the scanner and the high moisture content in sapwood. Instead, texture features in the central part of the logs were used in the classification.
Image processing methods were used to locate the knot-free areas and to segment the growth rings automatically. These methods were both simple and fast enough that they could be performed in parallel with the log scanning. A grey-level-gap-length method was used for texture feature extraction and a back propagation neural network was utilised for log classification.
The system was able to sort the logs into two different ring width classes with an 89% correct classification. The results show that automatic log sorting is feasible by X-ray scanning and image processing.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
R. C. Eberhart and R. W. Dobbins (Ed.), Neural Network PC Tools, Acad. Press, 1990.
A. Grönlund and S. Grundberg, Relations between Annual Ring Width and Different Properties for Individual Logs, 2nd IUFLO Workshop, Berg-en-Dal, South Africa, 1996.
G. T. Herman, Image Reconstuction from Projections, Academic Press, New York, 1980.
L. Jonsson, Recording Annual Ring Development Using Image Analysis, Workshop on Scanning Tech. and Image Proc. on Wood, Skellefteå Campus, Sweden, 1992.
L. A. Jozsa and G. R. Middleton, A Discussion of Wood Quality Attributes and Their Practical Implications, Foringtek Canada Corp. No. SP-34, 1994.
J. Kittler and J. Illingworth, Minimum Error Thresholding, Pattern Recognition, 19, 41–47, 1986.
L. O. Lindgren, Medical CAT-scanning, X-ray absorption coefficients, CT-numbers and their relation to wood density, Wood Sci. Technol. 25, 341–349, 1991.
H. Nyquist, Certain Topics in Telegraph Transmission Theory, AIEE Trans., 617–644, 1928.
X. Wang, F. Albregtsen and B. Foyn, Texture Analysis Using Grey Level Gap Length Matrix, in G. Borgefors (Ed.): “Theory and applications of image analysis II, Selected Papers from 9th SCIA”, World Scientific Publishing, Singapore, 65–78, 1995.
X. Wang, O. Hagman and S. Grundberg, Sorting pulpwood by X-ray Scanning, Proc. International Mechanical Pulping Conference, Sweden, 395–399, 1997.
X. Wang and K. R. Braaten, Growth rings and spruce pulpwood sorting, Nordic Pulp & Paper Research J., Vol. 3, 1997.
J. White and G. Rohrer, Image Thresholding for Optical Character Recognition and Other Applications Requiring Character Image Extraction, IBM J. of Research and Development, 27, 400–411, 1983.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Wang, X. (1997). Log classification by single X-ray scans using texture features from growth rings. In: Chin, R., Pong, TC. (eds) Computer Vision — ACCV'98. ACCV 1998. Lecture Notes in Computer Science, vol 1352. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-63931-4_207
Download citation
DOI: https://doi.org/10.1007/3-540-63931-4_207
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63931-2
Online ISBN: 978-3-540-69670-4
eBook Packages: Springer Book Archive