ABSTRACT
The taste of tea is one of the key indicators in the evaluation of its quality and is a key factor in its grading and market pricing. To objectively and digitally evaluate the taste quality of tea leaves, miniature near-infrared (NIR) spectroscopy and electronic tongue (ET) sensors are considered effective sensor signals for the characterization of the taste quality of tea leaves. This study used micro-NIR spectroscopy and ET sensors in combination with data fusion strategies and chemometric tools for the taste quality assessment and prediction of multiple grades of black tea. Using NIR features and ET sensor signals as fused information, the data optimization based on grey wolf optimization, ant colony optimization (ACO), particle swarm optimization, and non-dominated sorting genetic algorithm II were employed as modeling features, combined with support vector machine (SVM), extreme learning machine and K-nearest neighbor algorithm to build the classification models. The results obtained showed that the ACO-SVM model had the highest classification accuracy with a discriminant rate of 93.56%. The overall results reveal that it is feasible to qualitatively distinguish black tea grades and categories by NIR spectroscopy and ET techniques.
Subject(s)
Taste , Tea , Tea/chemistry , Spectroscopy, Near-Infrared/methods , Electronic Nose , Algorithms , Support Vector MachineABSTRACT
Fermentation is a key black tea processing step and makes an important contribution to quality formation. Current approaches to fermentation monitoring are costly or laboratory-based. Here, we first evaluated the potential of at-line computer vision for detecting fermentation quality in a tea factory. A self-built industrial camera was used to collect tea samples at various fermentation durations. The correlations of color variables that were extracted from the images with key quality indicators in the tea samples were verified. Subsequently, partial least-squares regression models based on the color variables showed high prediction accuracy with residual prediction deviation values of 4.13, 3.53, and 3.39 for catechins, theaflavins and chlorophylls, respectively. Finally, the spatial and temporal distributions of indicators during fermentation were mapped to visualize the fermentation quality. This study realized low-cost, at-line and real-time detection for black tea fermentation, which provides technical support for the industrial and intelligent production of black tea.
Subject(s)
Camellia sinensis , Tea , Fermentation , Least-Squares AnalysisABSTRACT
BACKGROUND: High-quality tea requires leaves of similar size and tenderness. The grade of the fresh leaves determines the quality of the tea. The automated classification of fresh tea leaves improves resource utilization and reduces manual picking costs. The present study proposes a method based on an improved genetic algorithm for identifying fresh tea leaves in high-speed parabolic motion using the phenotypic characteristics of the leaves. During parabolic flight, light is transmitted through the tea leaves, and six types of fresh tea leaves can be quickly identified by a camera. RESULTS: The influence of combinations of morphology, color, and custom corner-point morphological features on the classification results were investigated, and the necessary dimensionality of the model was tested. After feature selection and combination, the classification performance of the Naive Bayes, k-nearest neighbor, and support vector machine algorithms were compared. The recognition time of Naive Bayes was the shortest, whereas the accuracy of support vector machine had the best classification accuracy at approximately 97%. The support vector machine algorithm with only three feature dimensions (equivalent diameter, circularity, and skeleton endpoints) can meet production requirements with an accuracy rate reaching 92.5%. The proposed algorithm was tested by using the Swedish leaf and Flavia data sets, on which it achieved accuracies of 99.57% and 99.44%, respectively, demonstrating the flexibility and efficiency of the recognition scheme detailed in the present study. CONCLUSION: This research provides an efficient tea leaves recognition system that can be applied to production lines to reduce manual picking costs. © 2022 Society of Chemical Industry.
Subject(s)
Algorithms , Support Vector Machine , Bayes Theorem , Plant Leaves , TeaABSTRACT
Geraniol is a potent tea odorant and exists mainly as geranyl glycoside in Camellia sinensis. Understanding the mechanisms of geraniol biosynthesis at molecular levels in tea plants is of great importance for practical improvement of tea aroma. In this study, geraniol and its glycosides from tea plants were examined using liquid chromatography coupled with mass spectrometry. Two candidate geraniol synthase (GES) genes (CsTPS) and two Nudix hydrolase genes (CsNUDX1-cyto and CsNUDX1-chlo) from the tea genome were functionally investigated through gene transcription manipulation and gene chemical product analyses. Our data showed that in tea leaves, levels of geranyl ß-primeveroside were dramatically higher than those of geranyl ß-glucoside, while free geraniol was undetectable in this study. A tempo-spatial variation of geranyl ß-primeveroside abundance in tea plants existed, with high levels in young and green tissues and low levels in mature or non-green tissues. Cytosolic CsNUDX1-cyto showed higher hydrolysis activity of geranyl-pyrophosphate to geranyl-monophosphate (GP) in vitro than did chloroplastidial CsNUDX1-chlo. A transgenic study revealed that expression of CsNUDX1-cyto resulted in significantly more geranyl ß-primeveroside in transgenic Nicotiana benthamiana compared with non-transgenic wild-type, whereas expression of CsNUDX1-chlo had no effect. An antisense oligo-deoxynucleotide study confirmed that suppression of CsNUDX1-cyto transcription in tea shoots led to a significant decrease in geranyl ß-primeveroside abundance. Additionally, CsNUDX1-cyto transcript levels and geranyl ß-primeveroside abundances shared the same tempo-spatial patterns in different organs in the tea cultivar "Shucha Zao," indicating that CsNUDX1-cyto is important for geranyl ß-primeveroside formation in tea plants. Results also suggested that neither of the two candidate GES genes in tea plants did not function as GES in transgenic N. benthamiana. All our data indicated that CsNUDX1-cyto is involved in geranyl ß-primeveroside production in tea plants. Our speculation about possible conversion from the chemical product of CsNUDX1-cyto to geranyl ß-primeveroside in plants was also discussed.
ABSTRACT
Camellia plants include more than 200 species of great diversity and immense economic, ornamental, and cultural values. We sequenced the transcriptomes of 116 Camellia plants from almost all sections of the genus Camellia. We constructed a pan-transcriptome of Camellia plants with 89 394 gene families and then resolved the phylogeny of genus Camellia based on 405 high-quality low-copy core genes. Most of the inferred relationships are well supported by multiple nuclear gene trees and morphological traits. We provide strong evidence that Camellia plants shared a recent whole genome duplication event, followed by large expansions of transcription factor families associated with stress resistance and secondary metabolism. Secondary metabolites, particularly those associated with tea quality such as catechins and caffeine, were preferentially heavily accumulated in the Camellia plants from section Thea. We thoroughly examined the expression patterns of hundreds of genes associated with tea quality, and found that some of them exhibited significantly high expression and correlations with secondary metabolite accumulations in Thea species. We also released a web-accessible database for efficient retrieval of Camellia transcriptomes. The reported transcriptome sequences and obtained novel findings will facilitate the efficient conservation and utilization of Camellia germplasm towards a breeding program for cultivated tea, camellia, and oil-tea plants.
Subject(s)
Camellia , Camellia/genetics , Camellia/metabolism , Phylogeny , Plant Breeding , Tea/metabolism , Transcriptome/geneticsABSTRACT
Ti-CFRP-Ti laminated stacks have been widely used in aviation, aerospace, shipbuilding and other industries, owing to its excellent physical and electrochemical properties. However, chip blockages occur easily when drilling into Ti-CFRP-Ti laminated stacks, resulting in a rapid rise of drilling temperature and an increase of axial drilling force, which may lead to the intensification of tool wear and a decline of drilling quality. Cutting force signals can effectively reflect the drilling process and tool condition, however, the traditional plate dynamometer is typically difficult in realizing the follow-up online measurement. Therefore, an intelligent tool holder system for real-time sensing of the cutting force is developed and constructed in this paper, and the variable parameter drilling method of Ti-CFRP-Ti laminated stacks is studied on this basis. Firstly, an intelligent tool holder system with high flexibility and adaptability is designed; Secondly, a cutting force signal processing method based on compressed sensing (CS) theory is proposed to solve the problem of high-frequency signal transmission; Lastly, the drilling experiment of Ti-CFRP-Ti laminated stacks is carried out based on the intelligent tool holder system, and the drilling parameters are optimized using a compromise programming approach and analytic hierarchy process (AHP). The comparison of results show that the optimized drilling parameters can effectively reduce the hole wall surface roughness and improve the drilling efficiency while ensuring a small axial force.
ABSTRACT
Withering is one of the most critical steps in the processing of black tea. The degree of withering affects the aroma quality of the finished tea. In this study, we used a pH indicator-based colorimetric sensor array in combination with hyperspectral imaging to intelligently evaluate the withering degree. After analyzing the difference between images taken before and after the reaction of pH indicators with withered leaves, six pH indicators were selected to build a sensor array. Then, the hyperspectral image of each pH indicator was obtained at wavelengths between 400 and 1000 nm. Nonlinear support vector machine (SVM) and least-squares (LS) SVM models were established to determine the degree of withering. Results revealed that the spectral information from single pH indicator failed to accurately evaluate the withering degree. The LS-SVM model achieved satisfactory discriminant results with the low-level data fusion of six pH indicators followed by principal component analysis for dimensionality reduction. The optimal model yielded accuracies of 93.75% and 90.00% for the calibration and prediction sets, respectively. The results indicated that colorimetric sensor array in combination with hyperspectral imaging can effectively determine the withering degree, thus providing a novel method for the intelligent processing of food and tea.
Subject(s)
Camellia sinensis , Tea , Hydrogen-Ion Concentration , Hyperspectral Imaging , Least-Squares Analysis , Support Vector MachineABSTRACT
Rapid monitoring of fermentation quality has been the key to realizing the intelligent processing of black tea. In our study, mixing ratios, sensing array components and reaction times were optimized before an optimal solution phase colorimetric sensor array was constructed. The characteristic spectral information of the array was obtained by UV-visible spectroscopy and subsequently combined with machine learning algorithms to construct a black tea fermentation quality evaluation model. The competitive adaptive reweighting algorithms (CARS)-support vector machine model discriminated the black tea fermentation degree with 100% accuracy. For quantification of catechins and four theaflavins (TF, TFDG, TF-3-G, and TF-3'-G), the correlation coefficients of the CARS least square support vector machine model prediction set were 0.91, 0.86, 0.76, 0.72 and 0.79, respectively. The results obtained within 2 min enabled accurate monitoring of the fermentation quality of black tea, which provides a new method and idea for intelligent black tea processing.
Subject(s)
Camellia sinensis , Catechin , Catechin/analysis , Fermentation , Spectrophotometry, Ultraviolet , Spectroscopy, Near-Infrared , TeaABSTRACT
Methyl jasmonate (MeJA), a volatile organic compound, is a principal flowery aromatic compound in tea. During the processing of black tea, MeJA is produced by jasmonic acid carboxyl methyltransferase (JMT) of the jasmonic acid (JA) substrate, forming a specific floral fragrance. CsJMT was cloned from tea leaves; the three-dimensional structure of CsJMT was predicted. Enzyme activity was identified, and protein purification was investigated. Site-directed deletions revealed that N-10, S-22, and Q-25 residues in the beginning amino acids played a key functional role in enzyme activity. The expression patterns of CsJMT in tea organs differed; the highest expression of CsJMT was observed in the fermentation process of black tea. These results aid in further understanding the synthesis of MeJA during black tea processing, which is crucial for improving black tea quality using specific fragrances and could be applied to the aromatic compound regulation and tea breeding improvement in further studies.
Subject(s)
Odorants , Tea , Acetates , Cyclopentanes , Methyltransferases , Oxylipins , Plant BreedingABSTRACT
Keemun black tea is classified into 7 grades according to the difference in its quality. The appearance and flavour are crucial indicators of its quality. This research demonstrates a rapid grading method of jointly using near-infrared reflectance spectroscopy (NIRS) and computer vision systems (CVS) to evaluate the flavour and appearance quality of tea. A Bruker MPA Fourier Transform near-infrared spectrometer was used to record the spectrum of samples. A computer vision system was used to capture the image of tea leaves in an unobstructed manner. 80 tea samples for each grade were analyzed. The performance of four NIRS feature extraction methods (principal component analysis, local linear embedding, isometric feature mapping, and convolutional neural network (CNN)) was compared in this study. Histograms of six geometric features (leaf width, leaf length, leaf area, leaf perimeter, aspect ratio, and rectangularity) of different tea samples were used to describe their appearance. A feature-level fusion strategy was used to combine softmax and artificial neural networks (ANN) to classify NIRS and CVS features. The results indicated that for an individual NIRS signal, CNN achieved the highest classification accuracy with the softmax classification model. The histograms of the combined shape features indicated that when the softmax classification model was used, the classification accuracy was also higher than ANN. The fusion of NIRS and CVS features proved to be the optimal combination; the accuracy of calibration, validation and testing sets increased from 99.29%, 96.67% and 98.57% (when the optimal features from a single-sensor were used) to 100.00%, 99.29% and 100.00% (when features from multiple-sensors were used). This study revealed that the combination of NIRS and CVS features can be a useful strategy for classifying black tea samples of different grades.
Subject(s)
Camellia sinensis , Tea , Computers , Plant Leaves , Spectroscopy, Near-InfraredABSTRACT
The main objectives of the study are to understand and explore critical feature wavelengths of the obtained near-infrared (NIR) data relating to dianhong black tea quality categories, we propose a multi-variable selection strategy based on the variable space optimization from big to small which is the kernel idea of a variable combination of the improved genetic algorithm (IGA) and particle swarm optimization (PSO) in this study. A rapid description based on the NIR technology is implemented to assess black tea tenderness and rankings. First, 700 standard samples from dianhong black tea of seven quality classes are scanned using a NIR system. The raw spectra acquired are preprocessed by Savitzky-Golay (SG) filtering coupled with standard normal variate transformation (SNV). Then, the multi-variable selection algorithm (IGA-PSO) is applied to compare with the single method (the IGA and PSO) and search the optimal characteristic wavelengths. Finally, the identification models are developed using a decision tree (DT), partial least-squares discriminant analysis (PLS-DA), and support vector machine (SVM) based on different kernel functions combined with the effective features from the above variables screening paths for the discrimination of black tea quality. The results show that the IGA-PSO-SVM model with a radial basis function achieves the best predictive results with the correct discriminant rate (CDR) of 95.28% based on selected four characteristic variables in the prediction process. The overall results demonstrate that NIR combined with a multi-variable selection method can constitute a potential tool to understand the most important features involved in the evaluation of dianhong black tea quality helping the instrument manufacturers to achieve the development of low-cost and handheld NIR sensors.
Subject(s)
Camellia sinensis , Tea , Algorithms , Least-Squares Analysis , Spectroscopy, Near-Infrared , Support Vector MachineABSTRACT
Tea quality is generally assessed through panel sensory evaluation, which requires elaborate sample preparation steps. Here, a novel and low-cost evaluation method of using smartphone imaging coupled with micro-near-infrared (NIR) spectrometer based on digital light processing is proposed to classify the quality grades of Keemun black tea. RGB color information was obtained by Image J software, eight texture characteristics, including scheme, contrast, dissimilarity, entropy, correlation, second moment and variance, and homogeneity were obtained by ENVI software based on co - occurrence method from smartphone images, and spectral data were preprocessed with standard normal variate. A principal component analysis (PCA)-support vector machine (SVM) model was established to analyze the color, texture, and spectral data. Low-level and middle-level fusion strategies were introduced for analyzing the fusion data. The results indicated that the accuracy of the SVM model on mid-level data fusion (100.00%, 94.29% for calibration set and prediction set, respectively) was higher than that obtained for separate color (97.14%, 88.57%), texture (84.29%, 60%), spectrum (74.29%, 68.57%) evaluation, or low-level data fusion (88.57%, 82.86%). The best SVM model yielded satisfactory performance with 94.29% accuracy for the prediction sets. These results suggested that smartphone imaging coupled with micro-NIR spectroscopy is an effective and low-cost tool for evaluating tea quality.
Subject(s)
Camellia sinensis , Tea , Smartphone , Spectroscopy, Near-Infrared , Support Vector MachineABSTRACT
BACKGROUND: Tea (Camellia sinensis L) is a highly nutritious beverage with commercial value globally. However, it is at risk of economic fraud. This study aims to develop a powerful evaluation method to distinguish Chinese official Dianhong tea from various other categories, employing hyperspectral imaging (HSI) technology and chemometric algorithms. RESULTS: Two matrix statistical algorithms encompassing a gray-level co-occurrence matrix (GLCM) and a gradient co-occurrence matrix (GLGCM) are used to extract HSI texture data. Three novel spectral variable screening methods are utilized to select wavenumbers of near-infrared (NIR) spectra: iteratively retaining informative variables (IRIV), interval random frog, and variable combination population analysis. Feature fusion of image texture characteristics and spectra data are the eigenvectors for model building. Authentic classification models are constructed using the extreme learning machine approach and the least squares support vector machine (LSSVM) approach, coupling them with features from wavelength extraction techniques for assessing the quality of Dianhong black tea. The results demonstrate that the LSSVM model using fused data (IRIV + GLGCM) provides the best results and achieves a predictive precision of 99.57%. CONCLUSION: This study confirms that HSI coupled with LSSVM is effective in differentiating authentic Dianhong black tea samples. © 2020 Society of Chemical Industry.
Subject(s)
Camellia sinensis/chemistry , Hyperspectral Imaging/methods , Spectroscopy, Near-Infrared/methods , Tea/chemistry , Algorithms , Plant Leaves/chemistry , Quality ControlABSTRACT
Rapid and low-cost testing tools provide new methods for the evaluation of tea quality. In this study, a micro near-infrared (NIR) spectrometer was used for the qualitative and quantitative evaluation of tea. A total of 360 tea samples consisting of black, green, yellow, and oolong tea were collected from different countries. Chemometrics including linear partial least squares (PLS) regression, PLS discriminant analysis, and nonlinear radial basis function-support vector machine (RBF-SVM) were used. The RBF-SVM model achieved optimal discriminant performance for tea types with a correct classification rate of 98.33%. Wavelength selection of iteratively variable subset optimization (IVSO) exhibited considerable advantages in improving the predictive performance of catechin, caffeine, and theanine models. The IVSO-PLS regression models achieved satisfactory results for catechins and caffeine prediction, with Rp over 0.9, and RPD over 2.5. Thus, the study provided a portable and low-cost method for in-situ assessing tea quality.
Subject(s)
Food Analysis/instrumentation , Food Quality , Green Chemistry Technology/instrumentation , Spectroscopy, Near-Infrared/instrumentation , Tea/chemistry , Discriminant Analysis , Least-Squares Analysis , Support Vector MachineABSTRACT
Green tea adulterated with sugar and glutinous rice flour has an increased sensitivity to water, which affects the safety of the tea. A total of 475 samples of pure tea, sugar-adulterated tea, and glutinous-rice-flour-adulterated tea were prepared and scanned using micro near infrared spectroscopy (NIRS). The collected NIRS data were qualitatively and quantitatively detected by a multi-layer algorithm model. Principal component analysis indicated that the three sample groups had an obvious separation trend. The discriminate rate of the optimal qualitative model, namely support vector machine, was 97.47% for the prediction set. A total of three wavelength selection methods were used to improve the performances of partial least squares regression and support vector machine regression (SVR) models. The nonlinear SVR models based on characteristic wavelengths selected by iteratively retaining informative variables algorithm provided satisfactory results for the identification of sugar and glutinous rice flour adulteration. The correlation coefficients for prediction (Rp) were >0.94, and the residual prediction deviation were >3. The results indicated that smartphone-based micro NIRS can be effectively used to qualitatively and quantitatively analyze adulterants in green tea.
Subject(s)
Spectroscopy, Near-Infrared , Tea , Least-Squares Analysis , Quality Control , SmartphoneABSTRACT
While caffeine is one of the most important bioactive metabolites for tea as the most consumed non-alcohol beverage, its biosynthesis and catabolism in tea plants are still not fully understood. Here, we integrated purine alkaloid profiling and transcriptome analysis on shoot tips and roots fed with caffeine, theophylline, or theobromine to gain further understanding of caffeine biosynthesis and degradation. Shoot tips and roots easily took up and accumulated high concentrations of alkaloids, but roots showed much faster caffeine and theophylline degradation rates than shoot tips, which only degraded theophylline significantly but almost did not degrade caffeine. Clearly feedback inhibition on caffeine synthesis or inter-conversion between caffeine, theophylline, and theobromine, and 3-methylxanthine had been observed in alkaloids-fed shoot tips and roots, and these were also evidenced by significant repression of TCS and MXMT genes critical for caffeine biosynthesis. Among these responsively repressed genes, two highly expressed genes TCS-4 and TCS-8 were characterized for their enzyme activity. While we failed to detect TCS-4 activity, TCS-8 displayed N-methyltransferase activities towards multiple substrates, supporting the complex metabolic network in caffeine biosynthesis in tea plants since at least 13 TCS-like N-methyltransferase genes may function redundantly. This study provides new insight into complex metabolic networks of purine alkaloids in tea plants.
ABSTRACT
Caffeine and catechin are two main components of instant green tea, and are essential components of tea quality. This paper mainly focuses on the feasibility of rapidly determining instant green tea components by using a portable near infrared (NIR) spectrometer. The two main components (caffeine and catechin) were studied. In addition, the instrument performance levels of portable and benchtop NIR spectrometers were studied and compared. Quantitative models developed using portable and benchtop spectrometers for measuring caffeine, total catechins, and four individual catechins were established and compared. After preprocessing using standard normal variate (SNV), the Rp values of the caffeine, total catechins, (-)-epigallocatechin, (-)-epigallocatechin 3-gallate, (-)-epicatechin, and (-)-epicatechin gallate in the partial least squares models for a portable NIR spectrometer were 0.974, 0.962, 0.669, 0.945, 0.942 and 0.905, respectively. For a benchtop NIR spectrometer, Rp values were 0.993, 0.958, 0.883, 0.955, 0.966 and 0.936, respectively. Passing-Bablok regression method results indicated no significant differences between the two instruments. A genetic algorithm (GA) and the successive projections algorithm (SPA) were used to screen the wavelength of the NIR spectrum and establish the model. The GA obtained more robust modeling results. This study concludes that the developed portable spectroscopy system combined with appropriate variable selection methods can be effectively used for rapid determination of caffeine, total catechins, and four individual catechins in instant green tea.
Subject(s)
Catechin , Tea , Caffeine/analysis , Catechin/analysis , Chromatography, High Pressure Liquid , Least-Squares Analysis , RefractometryABSTRACT
Wild tea plants, which are classified into different species in the section Thea of the genus Camellia, are widely distributed in southern China. Tea produced from these plants has a unique flavor, which is different from that of tea produced from tea cultivars. In this study, we performed a comparative analysis of morphology, phylogenetic relationships, and phenolic compound metabolism between two wild tea plants (Gujing and Siqiu) and a tea cultivar (Shuchazao). Siqiu and Gujing tea plants had similar morphological traits and could be phylogenetically classified into a same cluster, which was entirely separate from the cluster containing widely cultivated cultivars such as Camellia sinensis cv. Shuchazao. Combined metabolomic and transcriptome analyses revealed that UGT84a22 was highly expressed in Gujing leaves compared with Shuchazao and Siqiu leaves, which may lead to the high accumulation of galloylquinic acid in Gujing leaves. A 14-bp deletion spanning the -765-(-7â¯5â¯1) range in the F3'5'H promoter potentially led to low F3'5'H expression levels in Siqiu and Gujing tea plants, which severely disrupted the accumulation of trihydroxy flavonoids in Gujing and Siqiu tea leaves. The high astringency intensity in Gujing tea could be due to the high accumulation of proanthocyanidins and galloylquinic acid. The results of the present study may improve our understanding of the metabolic characteristics of each evolutionary group of species or varieties in the section Thea of the genus Camellia.
Subject(s)
Camellia sinensis , Camellia , China , Phylogeny , TeaABSTRACT
The evaluation of tea quality tended to be subjective and empirical by human panel tests currently. A convenient analytical approach without human involvement was developed for the quality assessment of tea with great significance. In this study, near-infrared hyperspectral imaging (HSI) combined with multiple decision tree methods was utilized as an objective analysis tool for delineating black tea quality and rank. Data fusion that integrated texture features based on gray-level co-occurrence matrix (GLCM) and short-wave near-infrared spectral features were as the target characteristic information for modeling. Three different types of supervised decision tree algorithms (fine tree, medium tree, and coarse tree) were proposed for the comparison of the modeling effect. The results indicated that the performance of models was enhanced by the multiple perception feature fusion. The fine tree model based on data fusion obtained the best predictive performance, and the correct classification rate (CCR) of evaluating black tea quality was 93.13% in the prediction process. This work demonstrated that HSI coupled with intelligence algorithms as a rapid and effective strategy could be successfully applied to accurately identify the rank quality of black tea.
Subject(s)
Cheminformatics/methods , Food Analysis/methods , Hyperspectral Imaging/methods , Spectroscopy, Near-Infrared/methods , Tea/chemistry , China , Food Analysis/statistics & numerical data , Food Quality , Hyperspectral Imaging/statistics & numerical data , Image Processing, Computer-Assisted/methods , Spectroscopy, Near-Infrared/statistics & numerical dataABSTRACT
Near-infrared (NIR) spectroscopy is an effective tool for analyzing components relevant to tea quality, especially catechins and caffeine. In this study, we predicted catechins and caffeine content in green and black tea, the main consumed tea types worldwide, by using a micro-NIR spectrometer connected to a smartphone. Local models were established separately for green and black tea samples, and these samples were combined to create global models. Different spectral preprocessing methods were combined with linear partial-least squares regression and nonlinear support vector machine regression (SVR) to obtain accurate models. Standard normal variate (SNV)-based SNV-SVR models exhibited accurate predictive performance for both catechins and caffeine. For the prediction of quality components of tea, the global models obtained results comparable to those of the local models. The optimal global models for catechins and caffeine were SNV-SVR and particle swarm optimization (PSO)-simplified SNV-PSO-SVR, which achieved the best predictive performance with correlation coefficients in prediction (Rp) of 0.98 and 0.93, root mean square errors in prediction of 9.83 and 2.71, and residual predictive deviations of 4.44 and 2.60, respectively. Therefore, the proposed low-price, compact, and portable micro-NIR spectrometer connected to smartphones is an effective tool for analyzing tea quality.