An optimum artificial neural network and a partial least square with discriminant analysis regression were developed and tested for accuracy in distinguishing two wood species by using near-infrared (NIR) spectrum. A mixed population of kiln-dried wood boards of western hemlock (Tsuga heterophylla (Raf.) Sarg.) and amabilis fir (Abies amabilis (Dougl.) Forbes) were scanned by NIR and then a random sub-set was water saturated under vacuum conditions and scanned again. This design aimed to capture the effect of moisture content above the fibre saturation point on the separation algorithms. Our results revealed that both modelling techniques can be effective tools for species recognition achieving correct identification of over 86% for fir and 94% for hemlock on either kiln-dried or fully saturated boards.
AbrahamA.2005. Artificial Neural Networks Handbook of Measuring System Design, edited by SydenhamPeter H. and ThornRichard. Stillwater, OK: John Wiley and Sons Ltd. ISBN: 0-470-02143-8.
2.
AlvesA., SantosA., RozenbergP., PâquesL. E., CharpentierJ. P., SchwanningerM. and RodriguesJ.2012. A common near infrared – based partial least squares regression model for the prediction of wood density of Pinus pinaster and Larix × eurolepis. Wood Science and Technology. 46: 157–175. doi: 10.1007/s00226-010-0383-x
3.
AnttiH., SjöströmM. and WallbäcksL.1996. Multivariate calibration models using NIR spectroscopy on pulp and paper industrial applications. Journal of Chemometrics. 10: 591–603. doi: 10.1002/(SICI)1099-128X(199609)10:5/6<591::AID-CEM474>3.0.CO;2-L
4.
AvramidisS., AuneJ. E. and OliveiraL.2004. Exploring pre-sorting and redrying strategies for Pacific coast hemlock square timbers. Journal of the Institute of Wood Science. 16(4): 189–198.
5.
BouslamtiM. A., IrleM. A., BelloncleC., SalvadorV., HulotS.; CaronB. and QannariE. M.2013. Classification of treated wood using Fourier transform near infrared spectroscopy and multivariate data analysis. International Wood Products Journal. 4(2): 116–121. doi: 10.1179/2042645312Y.0000000026
6.
CooperP. A., JeremicD., RadivojevicS., UngY. T. and LeblonB. 2011. Potential of near-infrared spectroscopy to characterize wood products. Canadian Journal of Forest Research. 41: 2150–2157. doi: 10.1139/x11-088
7.
FujimotoT., NumaT., KoboriH. and TsuchikawaS.2015. Visualisation of spatial distribution of moisture content and basic density using near-infrared hyperspectral imaging method in sugi (Cryptomeria japonica). International Wood Products Journal. 6(2): 46–48. doi: 10.1179/2042645314Y.0000000084
8.
GuY., ShouG., ZhangW. and ZhaoD.2014. Application of near infrared spectroscopy for discrimination of similar rare woods in the Chinese market. Journal of Near Infrared Spectroscopy. 22(6): 423–431. doi: 10.1255/jnirs.1136
9.
InagakiT., SchwanningerM., KatoR., KurataY., ThanapaseW., PuthsonP. and TsuchikawaS.2012. Eucalyptus camaldulensis density and fiber length estimated by near-infrared spectroscopy. Wood Science and Technology. 46: 143–155. doi: 10.1007/s00226-010-0379-6
10.
KroghA.2008. What are artificial neural networks?Nature Biotechnology. 26(2): 195–197. doi: 10.1038/nbt1386
11.
LazarescuC., PanagiotidisK. and AvramidisS. 2014. Color variation of stained wood products in association to near infrared spectroscopy scans. European Journal of Wood and Wood Products. 72: 81–85. doi: 10.1007/s00107-013-0758-4
12.
LevenbergK.1944. A method for the solution of certain problems in least squares. Quarterly of Applied Mathematics. 2: 164–168. doi: 10.1090/qam/10666
13.
MarquardtD.1963. An algorithm for least-squares estimation of nonlinear parameters. Journal of the Society for Industrial and Applied Mathematics. 11: 431–441. doi: 10.1137/0111030
14.
MilotaM. R. and WuQ.1997. Post-sorting of hem-fir: a mill study. Forest Products Journal. 47(2): 49–56.
15.
OliveiraL., ZhangY. and AvramidisS.1994. Effect of species and basic density on the drying characteristics of 4 × 4 hem-fir lumber, 45th Western Dry Kiln Association, Meeting Reno, Nevada, US: 35–61.
16.
PastoreT. C. M., BragaJ. W. B., CoradinV. T. R., MagalhaesW. L. E., OkinoE. Y. A., CamargosJ. A. A., MunizG. I. B., BressanO. A. and DavrieuxF.2011. Near infrared spectroscopy (NIRS) as a potential tool for monitoring trade of similar woods: discrimination of true mahogany, cedar, andiroba, and curupixa. Holz. 65: 73–80.
17.
RojasR.1996. Neural Networks, a Systematic Introduction. Berlin, Germany: Springer-Verlag,502.
18.
TounisE.2009. Investigation of NIR spectroscopy for identifying and sorting wood with respect to species, moisture content and weathering, MSc Thesis, Ryerson University, Toronto, Canada, 176p.
19.
TsuchikawaS. and KoboriH. 2015. A review of recent application of near infrared spectroscopy to wood science and technology. Journal of Wood Science. 61: 213–220. doi: 10.1007/s10086-015-1467-x
20.
WatanabeK., MansfieldS. D. and AvramidisS.2011. Application of near-infrared spectroscopy for moisture-based sorting of green hem-fir timber. Journal of Wood Science. 57: 288–294. doi: 10.1007/s10086-011-1181-2
21.
WeyerL. G. and LoS. C.2002. Spectra-structure Correlations in the Near-infrared. Handbook of Vibrational Spectroscopy. Chichester: John Wiley & Sons.
22.
WileyK. N.1978. Site index tables for western hemlock in the Pacific Northwest. Weyerhaeuser Forestry Paper No. 17. Western Forestry Research Center, Centralia, Wash, USA. 28 p.
23.
YangS. Y., HanY., ChangY. S., KimK. M., ChoiI. G., and YeoH.2013. Moisture content predction below and above fiber saturation point by partial least squares regression analysis on near infrared absorption spectra of Korean pine. Wood and Fiber Science. 45: 415–422.
24.
ZhangY., OliveiraL., and AvramidisS.1996. Drying characteristics of square hem-fir lumber as affected by species. Forest Products Journal. 46(2): 44–50.
25.
ZupanJ.1994. Introduction to ANNs. Acta Chimica Slovenica. 41(3): 327–352.
