Three different SVM classification models in Tea Oil FTIR Application Research in Adulteration Detection

Fourier transform infrared spectroscopy (FTIR), as a new type of rapid environmental detection technology, has attracted extensive attention from researchers. In this study, FTIR combined with stoichiometry was used to establish and study the qualitative detection model of tea oil-doped soybean oil and corn oil. Different spectrum preprocessing and feature value extraction were performed on the infrared spectra of 105 samples of tea oil and adulterated oil in the spectral band of 600-4000cm-1. The KS sample selection method was used to divide the training set and the test set, and the training set was used to construct Support Vector Machine (SVM) Classification model, select the best model construction method based on the accuracy of the test set. The results show that the SVM tea oil adulteration detection model with convolution smoothing ( SG ) spectral preprocessing and random forest ( RF ) feature extraction achieves the best accuracy of 0.93, which is simple, fast, and pollution-free for the market to detect adulterated tea oil. Provide technical reference. (paper)