A comprehensive classification of wood from thermogravimetric curves

Abstract Wood is one of the most complicated materials to be classified in different classes or species. In this paper, the thermogravimetric (TG) curves of 49 wood samples are used to classify them in 7 predetermined species. Different functional and multivariate statistical supervised classification methods are used for this task: a nonparametric Nadaraya–Watson kernel functional estimator (K-NFDA), using the complete TG curves, and multivariate supervised classification approaches, such as linear discriminant analysis (LDA), k Nearest Neighbors ( k -NN), Naive Bayes (NBC), Neural Networks (NN), and Support Vector Machines (SVM). Before applying the multivariate techniques, the TG curves are discretized using principal component analysis (PCA) or fitting a four-component generalized logistic model. The results show that the classical method of LDA using the logistic parameters had the best performance, although high correct classification percentages were also obtained with the rest of the approaches. The work is completed with a comprehensive simulation study, comparing the classification techniques in different scenarios. Artificial TG curves are generated using the logistic model and additional conclusions on wood classification are established. Due to the heterogeneity of wood, this simulation study is very useful to describe worst-case scenarios and to assess more accurately the proposed classification methodologies.

[1]  Brian D. Ripley,et al.  Pattern Recognition and Neural Networks , 1996 .

[2]  Michel Ferriol,et al.  Thermal degradation of poly(methyl methacrylate) (PMMA): modelling of DTG and TG curves , 2003 .

[3]  Bin Li,et al.  Classification of functional data: A segmentation approach , 2008, Comput. Stat. Data Anal..

[4]  L. Jelemenský,et al.  Kinetic study of wood chips decomposition by TGA , 2010 .

[5]  M. Jurado-Expósito,et al.  Multispectral classification of grass weeds and wheat (Triticum durum) using linear and nonparametric functional discriminant analysis and neural networks , 2008 .

[6]  Jae Won Lee,et al.  An extensive comparison of recent classification tools applied to microarray data , 2004, Comput. Stat. Data Anal..

[7]  D. Mohan,et al.  Pyrolysis of Wood/Biomass for Bio-oil: A Critical Review , 2006 .

[8]  S. T. Buckland,et al.  An Introduction to the Bootstrap. , 1994 .

[9]  Javier Tarrío-Saavedra,et al.  Classification of wood micrographs by image segmentation , 2011 .

[10]  Ian R. Lewis,et al.  Raman spectrometry and neural networks for the classification of wood types—1 , 1994 .

[11]  Marco Gamassi,et al.  A low-cost neural-based approach for wood types classification , 2009, 2009 IEEE International Conference on Computational Intelligence for Measurement Systems and Applications.

[12]  Oxana Ye. Rodionova,et al.  Hard and soft methods for prediction of antioxidants' activity based on the DSC measurements , 2005 .

[13]  Haiping Yang,et al.  Characteristics of hemicellulose, cellulose and lignin pyrolysis , 2007 .

[14]  R. Artiaga,et al.  Thermogravimetric analysis of wood, holocellulose, and lignin from five wood species , 2012, Journal of Thermal Analysis and Calorimetry.

[15]  Vincenzo Piuri,et al.  Design of an Automatic Wood Types Classification System by Using Fluorescence Spectra , 2010, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[16]  A. Riga,et al.  Thermogravimetric Analysis (TGA) , 2008 .

[17]  Frédéric Ferraty,et al.  High-dimensional data: a fascinating statistical challenge , 2010, J. Multivar. Anal..

[18]  Javier Tarrío-Saavedra,et al.  Functional nonparametric classification of wood species from thermal data , 2011 .

[19]  Klaus Hechenbichler,et al.  Weighted k-Nearest-Neighbor Techniques and Ordinal Classification , 2004 .

[20]  Shingo Tomita,et al.  An optimal orthonormal system for discriminant analysis , 1985, Pattern Recognit..

[21]  Yong Haur Tay,et al.  Rotational Invariant Wood Species Recognition through Wood Species Verification , 2009, 2009 First Asian Conference on Intelligent Information and Database Systems.

[22]  Arbee L. P. Chen,et al.  Content-Based Query Processing for Video Databases , 2000, IEEE Trans. Multim..

[23]  Torsten Fransson,et al.  Comparison of the pyrolysis behavior of lignins from different tree species. , 2009, Biotechnology advances.

[24]  R. Artiaga,et al.  Estimating the reversing and non‐reversing heat flow from standard DSC curves in the glass transition region , 2011 .

[25]  William N. Venables,et al.  Modern Applied Statistics with S , 2010 .

[26]  R. Lippmann.  Pattern classification using neural networks , 1989, IEEE Communications Magazine.

[27]  R. Jordan,et al.  Classification of wood species by neural network analysis of ultrasonic signals , 1998 .

[28]  Edward Y. Chang,et al.  Support Vector Machine Concept-Dependent Active Learning for Image Retrieval , 2005 .

[29]  Manuel Febrero-Bande,et al.  Statistical Computing in Functional Data Analysis: The R Package fda.usc , 2012 .

[30]  R. Alén,et al.  Formation of the main degradation compound groups from wood and its components during pyrolysis , 1996 .

[31]  Marzuki Khalid,et al.  DESIGN OF AN INTELLIGENT WOOD SPECIES RECOGNITION SYSTEM , 2008 .

[32]  C. Blasi,et al.  Thermogravimetric Analysis and Devolatilization Kinetics of Wood , 2002 .

[33]  A. F. Roberts.  A review of kinetics data for the pyrolysis of wood and related substances , 1970 .

[34]  David J. Spiegelhalter,et al.  Machine Learning, Neural and Statistical Classification , 2009 .

[35]  Frances Whinder,et al.  Wood identification of Dalbergia nigra (CITES Appendix I) using quantitative wood anatomy, principal components analysis and naive Bayes classification. , 2010, Annals of botany.

[36]  John A. Nelder,et al.  A Simplex Method for Function Minimization , 1965, Comput. J..

[37]  Tomas Brandtberg,et al.  Individual tree-based species classification in high spatial resolution aerial images of forests using fuzzy sets , 2002, Fuzzy Sets Syst..

[38]  Ruth E. Norton,et al.  Identifying Wood: Accurate Results With Simple Tools , 1990 .

[39]  Godfried T. Toussaint,et al.  Geometric proximity graphs for improving nearest neighbor methods in instance-based learning and data mining , 2005, Int. J. Comput. Geom. Appl..

[40]  János Csirik,et al.  On naive Bayes in speech recognition , 2005 .

[41]  H. Lohninger,et al.  Identification of archaeological and recent wood tar pitches using gas chromatography/mass spectrometry and pattern recognition , 1990 .

[42]  Frédéric Ferraty,et al.  Curves discrimination: a nonparametric functional approach , 2003, Comput. Stat. Data Anal..

[43]  W. Miltyk,et al.  Recognition of tablet content by chemometric processing of differential scanning calorimetry curves—An acetaminophen example , 2010 .

[44]  Henry W. Altland,et al.  Applied Functional Data Analysis , 2003, Technometrics.

[45]  Henning Bockhorn,et al.  A comparative kinetic study on the pyrolysis of three different wood species , 2003 .

[46]  Keinosuke Fukunaga,et al.  Introduction to Statistical Pattern Recognition , 1972 .

[47]  R. Fisher.  THE USE OF MULTIPLE MEASUREMENTS IN TAXONOMIC PROBLEMS , 1936 .

[48]  Paolo Camorani,et al.  A Classification Method for Wood Types using Fluorescence Spectra , 2008, 2008 IEEE Instrumentation and Measurement Technology Conference.

[49]  R. Brereton,et al.  Pattern recognition for the analysis of polymeric materials. , 2006, The Analyst.

[50]  Daphne Koller,et al.  Support Vector Machine Active Learning with Applications to Text Classification , 2000, J. Mach. Learn. Res..

[51]  Igor Kononenko,et al.  Machine learning for medical diagnosis: history, state of the art and perspective , 2001, Artif. Intell. Medicine.

[52]  Ron Wehrens,et al.  Chemometrics with R: Multivariate Data Analysis in the Natural Sciences and Life Sciences , 2011 .

[53]  Kurt Hornik,et al.  Support Vector Machines in R , 2006 .

[54]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .