Study on Support Vector Machine Combined with Infrared Spectroscopy for Timber Species Identification

DOI

10.2991/icfcce-14.2014.26How to use a DOI?

Keywords

Timber Identification, Support Vector Machine, Infrared Spectroscopy, Cluster Analysis, Bayes Discriminant

Abstract

Via infrared spectroscopy (IR) combined with support vector machine (SVM), the study on timber species identification was carried on. Ten kinds of precious timber were used as experimental materials; each timber picked three sets of samples. The corresponding spectrum was recorded by infrared spectrometer. The spectral data was pretreated by baseline correction and dimensionality reduction. Radial basis function ( RBF )was selected as kernel function , and RBF coefficient ( ) was 0.01.As for cross-validation ,the model of timber species identification was respectively established by the adjustment of the training set and test set, the discriminant accuracy rate of three models were 70%, 80%, and 100 %. The optimal model was compared with the model of Cluster analysis and Bayes discriminant, which indicated that the SVM- infrared spectroscopy technology has better prediction results and certain research value for the development of the timber species identification.

Copyright

© 2014, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yi-Dan Sun
AU  - Jun-Yi He
AU  - Mei-Hua Wu
AU  - Jing-Jing Zheng
AU  - Yuan Gao
AU  - Xue-Shun Wang
PY  - 2014/03
DA  - 2014/03
TI  - Study on Support Vector Machine Combined with Infrared Spectroscopy for Timber Species Identification
BT  - Proceedings of the 2014 International Conference on Future Computer and Communication Engineering
PB  - Atlantis Press
SP  - 109
EP  - 112
SN  - 1951-6851
UR  - https://doi.org/10.2991/icfcce-14.2014.26
DO  - 10.2991/icfcce-14.2014.26
ID  - Sun2014/03
ER  -