Discrimination Between Goose Down and Duck Down by Fourier Transform Near-Infrared Spectroscopy

Y. Zhou,H. R. Xu,Y. B. Ying

DOI: https://doi.org/10.13031/2013.25229

2008-01-01

Transactions of the ASABE

Abstract:The percentage of duck down in goose down products is an important index in international trade. The traditional method for distinguishing between goose down and duck down uses a microscope and visual appraisal. However there are some problems with this method, such as limited samples and low accuracy. In this study, all samples (n=140) were divided into four groups, according to different proportions of goose down and duck down. A near-infrared (NIR) method combined with chemometrics was used to discriminate among them. A spectrum of each sample was acquired by a NEXUS intelligent FT-IR spectrometer using diffuse reflectance mode. Chemometrics methods, including principal component analysis (PCA) and discriminant partial least squares (DPLS), were used to analyze the spectra obtained. Various pre-treatments such as multiplicative scatter correction (MSC) and derivatives were also used to improve the model. PCA was used to distinguish between pure goose down and pure duck down. Results of PCA show that samples were divided completely into two groups without overlap except for one outlier The DPLS method was used to distinguish among the four groups. The DPLS method based on the derivative could increase the predictive precision of the model (r-values increase from 0.528 to 0.999), but the robustness of model was not satisfactory. Pre-treatment with MSC could improve the model in this aspect. The result shows that NIR is a fast and effective method for discrimination between goose down and duck down, even in mixtures.