Reliable Identification of Oolong Tea Species: Nondestructive Testing Classification Based on Fluorescence Hyperspectral Technology and Machine Learning

Abstract

A rapid and nondestructive tea classification method is of great significance in today’s research. This study uses fluorescence hyperspectral technology and machine learning to distinguish Oolong tea by analyzing the spectral features of tea in the wavelength ranging from 475 to 1100 nm. The spectral data are preprocessed by multivariate scattering correction (MSC) and standard normal variable (SNV), which can effectively reduce the impact of baseline drift and tilt. Then principal component analysis (PCA) and t-distribution random neighborhood embedding (t-SNE) are adopted for feature dimensionality reduction and visual display. Random Forest-Recursive Feature Elimination (RF-RFE) is used for feature selection. Decision Tree (DT), Random Forest Classification (RFC), K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) are used to establish the classification model. The results show that MSC-RF-RFE-SVM is the best model for the classification of Oolong tea in which the accuracy of the training set and test set is 100% and 98.73%, respectively. It can be concluded that fluorescence hyperspectral technology and machine learning are feasible to classify Oolong tea.

The Rapid Non-Destructive Detection of Adulteration and Its Degree of Tieguanyin by Fluorescence Hyperspectral Technology

“…Moreover, due to its strong growth adaptability, Ben tea's trading value is less than Tie's. Secondly, according to figures of PCA and t-SNE when reduced to 2 and 3 dimensions, it can be seen that there were lots of intersections between Tie and Ben, as is described by Hu et al [1]. In summary, using Ben as an adulterated tea was useful to this study.…”

“…Before analyzing the fluorescence spectra, it is very important to make a reasonable choice of the region of interest, which is directly related to the quality of the extracted data. In this experiment, the spectral data of ROI were extracted through ENVI 5.3 [1]. A rectangular region of interest around the tea sample was selected, and the mean value of the region value was taken as the sample spectral data.…”

“…As for the fluorescence characteristics of tea, as a plant, different substances in tea will absorb light in different bands and emit fluorescence in different bands. The previous study [1,30] mentioned the use of fluorescence hyperspectral technology to classify the varieties and grades of tea, which proved the feasibility of fluorescence hyperspectral technology in tea detection. In this study, FHSI is used to detect the degree of tea adulteration to further improve the application of fluorescence hyperspectral technology in tea detection.…”

“…Tea is a favorite health drink and one of the most widely used beverages today [1], which contains a range of amino acids, minerals, and vitamins [2]. Research has shown that tea has certain effects on health care and pharmacological impacts [3].…”

“…The above spectroscopic techniques greatly improve the efficiency of detection, but they suffer from some problems, such as the peak value they detect is not obvious, too much noise interference, and the experimental detection speed is slow [1]. To address these issues, fluorescence hyperspectral imaging technology (FHSI), which obtains spectral and spatial information simultaneously, can be used.…”