Annals of Forest Science

, 74:30

Automated classification of wood transverse cross-section micro-imagery from 77 commercial Central-African timber species

  • Núbia Rosa da Silva
  • Maaike De Ridder
  • Jan M. Baetens
  • Jan Van den Bulcke
  • Mélissa Rousseau
  • Odemir Martinez Bruno
  • Hans Beeckman
  • Joris Van Acker
  • Bernard De Baets
Original Paper

Abstract

Key message

Pattern recognition has become an important tool to aid in the identification and classification of timber species. In this context, the focus of this work is the classification of wood species using texture characteristics of transverse cross-section images obtained by microscopy. The results show that this approach is robust and promising.

Context

Considering the lack of automated methods for wood species classification, machine vision based on pattern recognition might offer a feasible and attractive solution because it is less dependent on expert knowledge, while existing databases containing high-quality microscopy images can be exploited.

Aims

This work focuses on the automated classification of 1221 micro-images originating from 77 commercial timber species from the Democratic Republic of Congo.

Methods

Microscopic images of transverse cross-sections of all wood species are taken in a standardized way using a magnification of 25 ×. The images are represented as texture feature vectors extracted using local phase quantization or local binary patterns and classified by a nearest neighbor classifier according to a triplet of labels (species, genus, family).

Results

The classification combining both local phase quantization and linear discriminant analysis results in an average success rate of approximately 88% at species level, 89% at genus level and 90% at family level. The success rate of the classification method is remarkably high. More than 50% of the species are never misclassified or only once. The success rate is increasing from the species, over the genus to the family level. Quite often, pattern recognition results can be explained anatomically. Species with a high success rate show diagnostic features in the images used, whereas species with a low success rate often have distinctive anatomical features at other microscopic magnifications or orientations than those used in our approach.

Conclusion

This work demonstrates the potential of a semi-automated classification by resorting to pattern recognition. Semi-automated systems like this could become valuable tools complementing conventional wood identification.

Keywords

Commercial timber species Democratic Republic of Congo Image analysis Pattern recognition Transverse cross-section Wood anatomy 

References

  1. Aha DW, Kibler D, Albert MK (1991) Instance-based learning algorithms. Mach Learn 6:37–66Google Scholar
  2. Beeckman H (2007) Collections of the RMCA. Wood. Royal Museum for Central Africa, TervurenGoogle Scholar
  3. Beeckman H (2016) Wood anatomy and trait-based ecology. IAWA J 37:127–151CrossRefGoogle Scholar
  4. Bremananth R, Nithya B, Saipriya R (2009) Wood species recognition system. World Acad Sci Eng Technol 52:873–879Google Scholar
  5. Cavalin PR, Kapp MN, Martins J, Oliveira LES (2013) A multiple feature vector framework for forest species recognition. In: Proceedings of the 28th annual ACM symposium on applied computing. ACM, New York, pp 16–20Google Scholar
  6. Cui K-P, Zhai X-R, Wang H-J (2013) A survey on wood recognition using machine vision. Adv For Lett 2:61–66Google Scholar
  7. Donkpegan AS, Doucet J-L, Migliore J, Duminil J, Dainou K, Piñeiro R, Wieringa JJ, Champluvier D, Hardy OJ (2017) Evolution in African tropical trees displaying ploidy-habitat association: the genus Afzelia (Leguminosae). Mol Phylogenet Evol 107:270–281CrossRefPubMedGoogle Scholar
  8. Donkpegan ASL, Hardy OJ, Lejeune P, Oumorou M, Daïnou K, Doucet J-L (2014) Un complexe d’espèces d’afzelia des forêts africaines d’intérêt économique et écologique (synthèse bibliographique). BASE 18:233–246Google Scholar
  9. Ebert DS (1994) Texturing and modeling: a procedural approach. Academic Press, CambridgeGoogle Scholar
  10. Filho PLP, Oliveira LS, Nisgoski S, Britto AS Jr (2014) Forest species recognition using macroscopic images. Mach Vis Appl 25:1019–1031CrossRefGoogle Scholar
  11. Fisher RA (1936) The use of multiple measurements in taxonomic problems. Ann Eugen 7:179–188CrossRefGoogle Scholar
  12. Giraud B (1980) Correlation between wood anatomical characters in entandrophragma utile (meliaceae). IAWA J 1:73–75CrossRefGoogle Scholar
  13. Guang-Sheng C, Peng Z (2013) Wood cell recognition using geodesic active contour and principal component analysis. Optik - Int J Light Electron Opt 124:949–952CrossRefGoogle Scholar
  14. Gurau L, Timar MC, Porojan M, Ioras F (2013) Image processing method as a supporting tool for wood species identification. Wood Fiber Sci 45:303–313Google Scholar
  15. Hanssen F, Wischnewski N, Moreth U, Magel EA (2011) Molecular identification of Fitzroya cupressoides, Sequoia sempervirens, and Thuja plicata wood using taxon-specific rDNA-ITS primers. IAWA J 32:273–284CrossRefGoogle Scholar
  16. Hastie T, Tibshirani R, Friedman J (2001) The elements of statistical learning. Springer Series in Statistics. Springer New York Inc., New YorkCrossRefGoogle Scholar
  17. Hermanson JC, Wiedenhoeft AC (2011) A brief review of machine vision in the context of automated wood identification systems. IAWA J 32:233–250CrossRefGoogle Scholar
  18. ITTO: International Tropical Timber Organization (2014) Annual review and assessment of the world timber situationGoogle Scholar
  19. Johnson A, Laestadius L (2011) New laws, new needs: the role of wood science in global policy efforts to reduce illegal logging and associated trade. IAWA J 32:125–136CrossRefGoogle Scholar
  20. Jolliffe IT (2002) Principal component analysis. Springer, New YorkGoogle Scholar
  21. Khairuddin ASM, Khalid M, Yusof R (2011) Using two stage classification for improved tropical wood species recognition system. In: Tsihrintzis G, Virvou M, Jain L, Howlett R (eds) Intelligent interactive multimedia systems and services, volume 11 of smart innovation, systems and technologies. Springer, Berlin, pp 305–314CrossRefGoogle Scholar
  22. Khalid M, Yusof R, Khairuddin ASM (2011) Improved tropical wood species recognition system based on multi-feature extractor and classifier. Int J Electr Comput Energ Electron Commun Eng 5:1490–1496Google Scholar
  23. Latham R, Ricklefs RE (1993) Continental comparisons of temperate-zone tree species diversity. In: Ricklefs R E, Schluter D (eds) Species diversity in ecological communities: historical and geographical perspectives. University of Chicago Press, Chicago, pp 294–314Google Scholar
  24. Lu J, Plataniotis K, Venetsanopoulos A (2005) Regularization studies of linear discriminant analysis in small sample size scenarios with application to face recognition. Pattern Recognit Lett 26:181–191CrossRefGoogle Scholar
  25. Mallik A, Tarrío-Saavedra J, Francisco-Fernández M, Naya S (2011) Classification of wood micrographs by image segmentation. Chemom Intell Lab Syst 107:351–362CrossRefGoogle Scholar
  26. Martins J, Oliveira LS, Sabourin R (2012) Combining textural descriptors for forest species recognition. In: Proceedings of the IECON 2012—38th annual conference of the IEEE Industrial Electronics Society, pp 1483–1488Google Scholar
  27. Martins J, Oliveira L S, Nisgoski S, Sabourin R (2013) A database for automatic classification of forest species. Mach Vis Appl 24:567–578CrossRefGoogle Scholar
  28. Musah RA, Espinoza EO, Cody RB, Lesiak AD, Christensen ED, Moore HE, Maleknia S, Drijfhout FP (2015) A high throughput ambient mass spectrometric approach to species identification and classification from chemical fingerprint signatures. Sci Rep 5:11520 EPCrossRefGoogle Scholar
  29. Nithaniyal S, Newmaster SG, Ragupathy S, Krishnamoorthy D, Vassou SL, Parani M (2014) DNA barcode authentication of wood samples of threatened and commercial timber trees within the tropical dry evergreen forest of india. PLoS One 9:e107669CrossRefPubMedPubMedCentralGoogle Scholar
  30. Ojala T, Pietikäinen M, Harwood D (1996) A comparative study of texture measures with classification based on featured distributions. Pattern Recognit 29:51–59CrossRefGoogle Scholar
  31. Ojala T, Pietikäinen M, Mäenpää T (2001) A generalized local binary pattern operator for multiresolution gray scale and rotation invariant texture classification. In: Proc 2nd international conference on advances in pattern recognition. Springer, London, pp 397–406Google Scholar
  32. Ojala T, Pietikäinen M, Mäenpää T (2002) Multiresolution gray-scale and rotation invariant texture classification with local binary patterns. IEEE Trans Pattern Anal Mach Intell 24:971–987CrossRefGoogle Scholar
  33. Ojansivu V, Heikkilä J (2008). In: Elmoataz A, Lezoray O, Nouboud F, Mammass D (eds) Image and signal Processing, volume of 5099 of lecture notes in computer science. Springer, Berlin, pp 236–243Google Scholar
  34. Ojansivu V, Rahtu E, Heikkila J (2008) Rotation invariant local phase quantization for blur insensitive texture analysis. In: Proceedings of the 19th international conference on pattern recognition, pp 1–4Google Scholar
  35. Ruffinatto F, Crivellaro A, Wiedenhoeft AC (2015) Review of macroscopic features for hardwood and softwood identification and a proposal for a new character list. IAWA J 36:208–241CrossRefGoogle Scholar
  36. Slik JWF, Arroyo-Rodríguez V, Aiba S-I, Alvarez-Loayza P, Alves LF, Ashton P, Balvanera P, Bastian ML, Bellingham PJ, van den Berg E, Bernacci L, da Conceição Bispo P, Blanc L, Böhning-Gaese K, Boeckx P, Bongers F, Boyle B, Bradford M, Brearley FQ, Breuer-Ndoundou Hockemba M, Bunyavejchewin S, Calderado Leal Matos D, Castillo-Santiago M, Catharino ELM, Chai S-L, Chen Y, Colwell RK, Chazdon RL, Clark C, Clark DB, Clark DA, Culmsee H, Damas K, Dattaraja HS, Dauby G, Davidar P, DeWalt SJ Doucet J-L, Duque A, Durigan G, Eichhorn KAO, Eisenlohr PV, Eler E, Ewango C, Farwig N, Feeley KJ, Ferreira L, Field R, de Oliveira Filho AT, Fletcher C, Forshed O, Franco G, Fredriksson G, Gillespie T, Gillet J-F, Amarnath G, Griffith DM, Grogan J, Gunatilleke N, Harris D, Harrison R, Hector A, Homeier J, Imai N, Itoh A, Jansen PA, Joly CA, de Jong BHJ, Kartawinata K, Kearsley E, Kelly DL, Kenfack D, Kessler M, Kitayama K, Kooyman R, Larney E, Laumonier Y, Laurance S, Laurance WF, Lawes MJ, Amaral ILD, Letcher SG, Lindsell J, Lu X, Mansor A, Marjokorpi A, Martin EH, Meilby H, Melo FPL, Metcalfe DJ, Medjibe VP, Metzger JP, Millet J, Mohandass D, Montero JC, de Morisson Valeriano M, Mugerwa B, Nagamasu H, Nilus R, Ochoa-Gaona S, Onrizal, Page N, Parolin P, Parren M, Parthasarathy N, Paudel E, Permana A, Piedade MTF, Pitman NCA, Poorter L, Poulsen AD, Poulsen J, Powers J, Prasad RC, Puyravaud J-P, Razafimahaimodison J-C, Reitsma J, dos Santos JR, Roberto Spironello W, Romero-Saltos H, Rovero F, Rozak AH, Ruokolainen K, Rutishauser E, Saiter F, Saner P, Santos BA, Santos F, Sarker SK, Satdichanh M, Schmitt CB, Schöngart J, Schulze M, Suganuma MS, Sheil D, da Silva Pinheiro E, Sist P, Stevart T, Sukumar R, Sun I-F, Sunderland T, Suresh HS, Suzuki E, Tabarelli M, Tang J, Targhetta N, Theilade I, Thomas DW, Tchouto P, Hurtado J, Valencia R, van Valkenburg JLCH, Van Do T, Vasquez R, Verbeeck H, Adekunle V, Vieira SA, Webb CO, Whitfeld T, Wich SA, Williams J, Wittmann F, Wöll H, Yang X, Adou Yao CY, Yap SL, Yoneda T, Zahawi RA, Zakaria R, Zang R, de Assis RL, Garcia Luize B, Venticinque EM (2015) An estimate of the number of tropical tree species. Proc Natl Acad Sci 112: 7472–7477Google Scholar
  37. Verbeeck H, Boeckx P, Steppe K (2011) Tropical forests: include congo basin. Nature 479:179–179CrossRefPubMedGoogle Scholar
  38. Wang H-J, Qi H-N, Wang X-F (2013a) A new Gabor based approach for wood recognition. Neurocomputing 116:192–200CrossRefGoogle Scholar
  39. Wang H-J, Zhang G-Q, Qi H-N (2013b) Wood recognition using image texture features. PLoS One 8:e76101CrossRefPubMedPubMedCentralGoogle Scholar
  40. Wheeler EA (2011) Inside wood—a web resource for hardwood anatomy. IAWA J 32:199–211CrossRefGoogle Scholar
  41. Wheeler EA, Baas P (1998) Wood identification—a review. IAWA J 19:241–264CrossRefGoogle Scholar
  42. Yu H, Cao J, Luo W, Liu Y (2009) Image retrieval of wood species by color, texture, and spatial information. In: Proceedings of the international conference on inform. automation, pp 1116– 1119Google Scholar
  43. Yusof R, Khalid M, Khairuddin ASM (2013a) Application of kernel-genetic algorithm as nonlinear feature selection in tropical wood species recognition system. Comput Electron Agric 93:68– 77CrossRefGoogle Scholar
  44. Yusof R, Khalid M, Khairuddin ASM (2013b) Fuzzy logic-based pre-classifier for tropical wood species recognition system. Mach Vis Appl 24:1589–1604CrossRefGoogle Scholar
  45. Zhao P (2013) Robust wood species recognition using variable color information. Optik - Int J Light Electron Opt 124:2833– 2836CrossRefGoogle Scholar
  46. Zhao P, Dou G, Chen G-S (2014a) Wood species identification using feature-level fusion scheme. Optik - Int J Light Electron Opt 125:1144–1148CrossRefGoogle Scholar
  47. Zhao P, Dou G, Chen G-S (2014b) Wood species identification using improved active shape model. Optik - Int J Light Electron Opt 125:5212–5217CrossRefGoogle Scholar

Copyright information

© INRA and Springer-Verlag France 2017

Authors and Affiliations

  • Núbia Rosa da Silva
    • 1
    • 2
    • 3
  • Maaike De Ridder
    • 4
    • 5
  • Jan M. Baetens
    • 6
  • Jan Van den Bulcke
    • 4
  • Mélissa Rousseau
    • 5
  • Odemir Martinez Bruno
    • 1
    • 2
  • Hans Beeckman
    • 5
  • Joris Van Acker
    • 4
  • Bernard De Baets
    • 6
  1. 1.Institute of Mathematics and Computer ScienceUniversity of São Paulo, USPSão CarlosBrazil
  2. 2.Scientific Computing Group, São Carlos Institute of PhysicsUniversity of São Paulo, USPSão CarlosBrazil
  3. 3.Institute of BiotechnologyFederal University of Goiás, UFGCatalãoBrazil
  4. 4.Department of Forest and Water Management, Faculty of Bioscience EngineeringGhent UniversityGhentBelgium
  5. 5.Royal Museum for Central Africa, Service of Wood BiologyTervurenBelgium
  6. 6.KERMIT, Department of Mathematical Modelling, Statistics and Bioinformatics, Faculty of Bioscience EngineeringGhent UniversityGhentBelgium

Personalised recommendations