Rapid Distinguish of Edible Oil Adulteration Using a Hyperspectral Spectroradiometer

The adulteration of edible oil is one of the widely existing oil quality problems. In this study, a hyperspectral spectroradiometer was utilised to discriminate sesame and rapeseed oil adulteration rapidly using soybean oil. The reflectance spectra of 350-2500 nm were obtained from different adulterated oil samples using the hyperspectral spectroradiometer. Support Vector Machine (SVM) and random forest (RF) were used for spectral analysis to classify adulteration types. And RF achieved an accuracy of 100%. Considering the redundant information of reflectance spectra, the optimal wavelengths were selected by Principal Component Analysis to increase processing speed and simplify the model for subsequent applications. With the spectra of eight selected wavelengths, the accuracy of SVM achieved 92.86%. These results indicated that a hyperspectral spectroradiometer can distinguish edible oil adulteration.