Identifying Dalbergia spp. wood with hyperspectral imaging technology

Abstract

To rapidly and nondestructively identify common Dalbergia spp. of wood on the market, wood from Dalbergia spp. (D. cochinchinensis, D. bariensis, D. oliveri, and D. retusa) was identified using hyperspectral imaging technology. The hyperspectral images were collected and the reflectance spectral from the region of interest were extracted from the images. Wavelengths from 955 to 1 642 nm were preprocessed by Savitsky-Golay smoothing(SG), standard normal variate(SNV), and Multiplicative Scatter Correction(MSC). Then, a partial least square-discriminant analysis(PLS-DA) and an extreme learning machine (ELM) were used to build discriminant models based on selected sensitive wavelengths using principal component analysis (PCA), regression coefficient (RC), and successive projections algorithm (SPA) from the preprocessed spectra. Results showed that for selected sensitive wavelengths using SPA from SG and MSC preprocessed spectra, ELM models obtained the best classification accuracy (100.0%) for both the calibration set and the prediction set. Thus, this study provided a new method to identify Dalbergia spp. wood rapidly and nondestructively.[Ch, 5 fig. 4 tab. 17 ref.]