Classification of ginseng according to plant species, geographical origin, and age using laser-induced breakdown spectroscopy and hyperspectral imaging

Abstract

Because of the increasing demand and consumption of ginseng products, rapid and effective technologies verifying the authenticity of ginseng are strongly needed. The constituents of ginseng differ by plant species, geographical origin, and age, thus resulting in different medicinal values and market prices. Traditional methods for the quality assessment of ginseng have generally been based on sensory analysis and wet chemistry, which are time-consuming and require specialized knowledge. In this work, we used laser-induced breakdown spectroscopy (LIBS) and hyperspectral imaging (HSI) combined with chemometrics to determine the plant species, geographical origin, and age of ginseng samples. Ginseng slices were pretreated into pressed pellets and then emission and reflectance spectra were acquired. Partial least-squares discriminant analysis (PLS-DA) was used to construct models based on spectra and label information. The classification accuracies of the spectroscopic measurements of ginseng by plant species, geographical origin, and age were greater than 93%, 94%, and 99%, respectively. Measurement fusion further improved the results of geographical origin classification by at least 2%. In addition, we selected the key wavelengths for the LIBS measurement and identified the chemical elements related to classification. The experimental results demonstrated the effectiveness of LIBS and HSI in the rapid assessment of ginseng authenticity.