Skip to main content
Log in

Identification of internal defect of sugar maple logs from CT images using supervised classification methods

Erkennung innerer Fehler in Stammabschnitten von Zuckerahorn mit vorwissensbasierter Auswertung von CT-Bildern

  • ORIGINALARBEITEN ORIGINALS
  • Published:
Holz als Roh- und Werkstoff Aims and scope Submit manuscript

Abstract

Sugar maple logs (Acer saccharum Marsh.) were scanned using X-ray medical scanner in order to develop a classification procedure for this type of imagery (CT images). The classification procedure was required in order to separate sapwood from colored heartwood, knots, rot and bark. Five logs coming from a freshly cut tree (group 1) and three logs sampled from a sawmill yard (group 2) were chosen for this purpose. Two parametric supervised classification algorithms, a minimum distance (MDC) and maximum likelihood (MLC) ones, were qualitatively and quantitatively tested and the resulting thematic maps were filtered by a 5×5 median filter. The classification accuracy was evaluated with confusion matrix and Kappa analyses. Sapwood is known to be the key factor determining sugar maple lumber value. The sapwood identification accuracy was found to be 98.6% (MDC) and 97.2% (MLC) for group 1 and 80.7% (MDC) and 81.8% (MLC) for group 2, respectively. Misclassification of defects occurred mainly between knots and colored heartwood. The overall accuracy of classification was about 83.1% (MDC) and 82.6% (MLC) for group 1 and 76.4 (MDC) and 78.0% (MLC) for group 2, respectively. The Kappa value from MDC and MLC was 0.622 and 0.624 for group 1 and 0.440 and 0.470 for group 2, respectively. These Kappa values indicate the existence of strong and moderate degree of conformity between the reference data and the classification procedure for groups 1 and 2 of logs, respectively. Both classifiers show no statistically significant differences in their capability of separation of sapwood from the other classes. Nevertheless, as MLC accuracy for colored heartwood is higher than MDC accuracy in logs without bark (normal situation in sawmills), MLC appears at this stage as the better alternative for analysing CT images of sugar maple logs.

Zusammenfassung

Am Beispiel von Zuckerahorn-Stammabschnitten (Acer saccarum Marsh.), die mit einem Computertomographen gescannt wurden, wurde ein Auswerteverfahren für die CT-Bilder entwickelt. Dieses Verfahren war erforderlich, um Splintholz von Farbkernholz, Ästen, Fäule und Rinde unterscheiden zu können. Dazu wurden fünf Stammabschnitte von frisch gefällten Bäumen (Gruppe 1) und drei Stammabschnitte aus einem Sägewerk (Gruppe 2) entnommen. Zwei vorwissensbasierte Auswerteverfahren, Minimum Distance (MDC) und Maximum-Likelihood (MLC), wurden sowohl qualitativ als auch quantitativ getestet. Die daraus resultierenden thematischen Karten wurden mit einem 5×5 Medianfilter gefiltert. Die Auswertegenauigkeit wurde anhand einer Konfusionsmatrix und mittels Kappa-Analysen bewertet. Splintholz ist der entscheidende Faktor, der den Wert von Zuckerahorn-Schnittholz bestimmt. Splintholz konnte in der Gruppe 1 mit einer Genauigkeit von 98.6% (MDC) und 97.2% (MLC) und in der Gruppe 2 mit 80.7% (MDC) und 81.8% (MLC) erkannt werden. Auswertefehler ergaben sich hauptsächlich bei der Unterscheidung zwischen Ästen und Farbkernholz. Insgesamt lag die Auswertegenauigkeit in der Gruppe 1 bei 83.1% (MDC) und 82.6% (MLC) und in der Gruppe 2 bei 76.4% (MDC) und 78.0% (MLC). Der Kappa-Wert von MDC und MLC lag in der Gruppe 1 bei 0.622 und 0.624 bzw. in der Gruppe 2 bei 0.440 und 0.470. Diese Kappa-Werte weisen auf einen hohen bzw. mittleren Übereinstimmungsgrad zwischen den Referenz- und den Auswertedaten für Stammabschnitte der Gruppen 1 bzw. 2 hin. Beide Verfahren unterscheiden sich statistisch nicht signifikant bezüglich ihrer Fähigkeit, Splintholz von den anderen Holzmerkmalen zu trennen. Dennoch erscheint zum gegenwärtigen Zeitpunkt MLC die bessere Alternative zur Auswertung von CT-Bildern von Zuckerahorn-Stammabschnitten zu sein, da MLC Farbkernholz in Stammabschnitten ohne Rinde (Normalsituation in Sägewerken) mit höherer Genauigkeit identifiziert als MOC.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  1. Bahandarkar SM, Faust TD, Tang M (1999) CATALOG: a system for detection and rendering of internal log defects using computer tomography. Mach Vis Appl 3:171–190

    Article  Google Scholar 

  2. Birkeland R, Han W (1991) Ultrasonic scanning for internal log defects. 4th international conference on scanning technology in the wood industry. Burlingame, California, USA, October 28–29

  3. Chang SJ, Olson JR, Wang PC (1989) NMR imaging of internal features in wood. For Prod J 39(6):43–49

    CAS  Google Scholar 

  4. Funt BV, Bryant EC (1987) Detection of internal log defects by automatic interpretation of computer tomography images. For Prod J 37(1):56–62

    Google Scholar 

  5. Gonzalez RC, Woods RE (2002) Digital image processing. Prentice Hall, Upper Saddle River, New Jersey, p 793

    Google Scholar 

  6. Jensen JR (1996) Introductory digital image processing: a remote sensing perspective. Prentice Hall, Upper Saddle River, New Jersey, p 318

    Google Scholar 

  7. Landis JR, Koch GC (1977) The measurement of observer agreement of categorical data. Biometrics 33:159–174

    Article  PubMed  CAS  Google Scholar 

  8. Li P, Abbot L, Schmoldt DL (1996) Automated analysis of CT images for the inspection of hardwood logs. Proceedings of the IEEE International Conference on Neural Networks, Washington DC, USA, pp 1744–1749

  9. Mather PM (1987) Computer processing of remotely sensed images, an introduction. Bury St Edmunds, Suffolk, p 352

    Google Scholar 

  10. Nordmark U (2002) Knot identification from CT images of young Pinus sylvestris saw logs using artificial neural networks. Scand J For Res 17:72–78

    Article  Google Scholar 

  11. Nordmark U (2003) Models of Knot and Log Geometry of Young Pinus sylvestris Sawlogs extracted from Computed Tomographic Images. Scand J For Res 18:168–175

    Article  Google Scholar 

  12. PCI user guide (1997) Volume I. Version 6.1. 50 West Wilmot street, Richmond Hill, Ontario, Canada

  13. Richards JA, Jia X (1999) Remote sensing digital image analysis an introduction. Springer-Verlag, Berlin Heidelberg New York, p 363

    Google Scholar 

  14. Rojas EG, Hernández RE, Condal A, Verret D, Beauregard R (2005) Exploration of the physical properties of internal characteristics of Sugar Maple logs and relationships with CT images. Wood Fiber Sci 37(4):591–604

    Google Scholar 

  15. Schmoldt DL, Li P, Abbott AL (1995) Log defect recognition using CT-images and neural net classifiers. 2nd International Workshop/Seminar on Scanning Technology and Image Processing on Wood. Skellefteå Sweden, Aug 14–16, pp 77–97

  16. Schmoldt DL, He J, Abbott AL (1998) A comparison of several artificial neural network classifiers for CT images of hardwood logs. Machine vision applications in industrial inspection VI. The International Society for Optical Engineering (SPIE), vol 3306, pp 34-43

  17. Schmoldt DL, He J, Abbott AL (2000) Automated labelling of log features in CT imagery of multiple hardwood species. Wood Fiber Sci 32(3):287–300

    CAS  Google Scholar 

  18. Story M, Congalton RG (1986) Accuracy assessment: a user’s perspective. Photogramm Eng Remote Sens 54(3):397–399

    Google Scholar 

  19. Taylor FW, Wagner Jr FG, Mc Millin CW (1984) Locating knots by industrial tomography-A feasibility study. For Prod J 34(5):42–46

    Google Scholar 

  20. Zhu D, Conners RW, Araman PA (1991) CT image processing for hardwood log inspection. Applications of digital image processing. The International Society for Optical Engineering (SPIE), vol 1567, pp 232–240

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Gerson Rojas.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Rojas, G., Condal, A., Beauregard, R. et al. Identification of internal defect of sugar maple logs from CT images using supervised classification methods. Holz Roh Werkst 64, 295–303 (2006). https://doi.org/10.1007/s00107-006-0105-0

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00107-006-0105-0

Keywords

Navigation