Identifying the geographical origins of white tea is of significance because the quality and price of white tea from different production areas vary largely from different growing environment and climatic conditions. In this study, we used near-infrared spectroscopy (NIRS) with white tea (n = 579) to produce models to discriminate these origins under different conditions. Continuous wavelet transform (CWT), min-max normalization (Minmax), multiplicative scattering correction (MSC) and standard normal variables (SNV) were used to preprocess the original spectra (OS). The approaches of principal component analysis (PCA), linear discriminant analysis (LDA) and successive projection algorithm (SPA) were used for features extraction. Subsequently, identification models of white tea from different provinces of China (DPC), different districts of Fujian Province (DDFP) and authenticity of Fuding white tea (AFWT) were established by K-nearest neighbors (KNN), random forest (RF) and support vector machine (SVM) algorithms. Among the established models, DPC-CWT-LDA-KNN, DDFP-OS-LDA-KNN and AFWT-OS-LDA-KNN have the best performances, with recognition accuracies of 88.97%, 93.88% and 97.96%, respectively; the area under curve (AUC) values were 0.85, 0.93 and 0.98, respectively. The research revealed that NIRS with machine learning algorithms can be an effective tool for the geographical origin traceability of white tea.

No abstract

Compressed white tea (CWT) is a reprocessed tea of white tea. Long-term storage has greatly changed its aroma characteristics, but the material basis and transformation mechanism of its unique aroma are still unclear. In this study, flavor wheel, headspace gas chromatography ion mobility spectroscopy, chemometrics, and microbiomics were applied to study the flavor evolution and important aroma components during long-term storage of CWT, and core functional bacteria were screened. During long-term storage, the aroma of CWT gradually changed from sweet, fruity and floral to stale flavor, woody and herbal. A total of 56 volatile organic compounds (VOCs) were identified, 54 of which were significantly differences during storage. The alcohols content was the highest during 1–5 years of storage, the esters content was the highest during 7–13 years of storage, and the aldehydes content was the highest during 16 years of storage. Twenty-nine VOCs were identified as important aroma components, which were significantly correlated with 6 aroma sub-attributes (P < 0.05). The functional prediction of bacterial community reminded that bacterial community could participate in the transformation of VOCs during storage of CWT. Twenty-four core functional bacteria were screened, which were significantly associated with 29 VOCs. Finally, 23 characteristic differential VOCs were excavated, which could be used to identify CWT in different storage years. Taken together, these findings provided new insights into the changes in aroma characteristics during storage of CWT and increased the understanding of the mechanism of characteristic aroma formation during storage.

No abstract

In order to solve the problem of calculating the instantaneous frequency by means of local mean decomposition, four methods were introduced such as Hilbert transform, Teager energy operator, direct quadrature and inverse cosine. By analysis of the advantages and disadvantages of various methods, the following conclusions could be made: first , the phenomenon of end swing by Hilbert transform was obvious; second, Teager energy operator method was easy to susceptible to signal form and noise, and the error was relatively large; third, inverse cosine method could be influenced by signal extreme points and brought in the mutations of instantaneous frequency. To solve the mutations of instantaneous frequency by inverse cosine method, an improved algorithm was proposed, which used cubic B-spline interpolation to obtain the instantaneous frequency of signal extreme points. It was shown from simulation and experimental results that the improved inverse cosine method could avoid the phenomenon of mutation and the error of instantaneous frequency was smaller.Index Terms -local mean decomposition, instantaneous frequency, inverse cosine, Teager energy operator.