NI-Raman spectroscopy combined with BP-Adaboost neural network for adulteration detection of soybean oil in camellia oil

To improve the accuracy of detecting adulteration of camellia oil, a real-time and rapid method using near-infrared Raman spectroscopy combined with BP-Adaboost neural network (BP-ANN) model was developed for the first time. Principal component analysis was used to reduce the dimension of the Raman spectral data of pure camellia oil, soybean oil, and camellia oil blended with soybean oil. Adaboost algorithm was used to optimize the back propagation neural network model for differentiating between authentic and adulterated camellia oil. A BP-ANN model was successfully established to detect the adulteration of camellia oil. This model obtained the value of prediction linearity greater than 0.999 and a mean square error less than 1%. Our results demonstrated that NIRS combined with BP-ANN model could be used as a rapid and precise approach for authentication and adulteration analysis of camellia oil.