A Framework for the Multi-Level Fusion of Electronic Nose and Electronic Tongue for Tea Quality Assessment

Are there any differences in the quality of high-mountain green tea before and after the first new leaves unfold? A comprehensive study based on E-sensors, whole metabolomics and sensory evaluation

High-mountain green tea, where the first new leaf hasn't yet unfurled, is prized for perceived superior quality, but this hasn't yet been verified by experimentation. Electronic sensors, whole metabolomics and sensory evaluation were employed to assess the quality of yymj (tea buds with a newly unfurled leaf) and qymj (tea buds without new leaves). The qymj proved to have significant advantages in aroma, color and shape, but still had some shortcomings in umami, bitterness and sourness. Differences in the content of volatile organic compounds (including alcohols, hydrocarbons and lipids) and nonvolatile organic compounds (flavonoids, amino acids, sugars, and phenolic acids) quality of high-mountain green teas with different maturity levels and provides well explained these quality differences. This study establishes a systematic approach to study the quality of high-mountain green tea at different maturity levels, and provides important reference information for consumers, governments and tea farmers.

Rapid Color Quality Evaluation of Needle-Shaped Green Tea Using Computer Vision System and Machine Learning Models

Color characteristics are a crucial indicator of green tea quality, particularly in needle-shaped green tea, and are predominantly evaluated through subjective sensory analysis. Thus, the necessity arises for an objective, precise, and efficient assessment methodology. In this study, 885 images from 157 samples, obtained through computer vision technology, were used to predict sensory evaluation results based on the color features of the images. Three machine learning methods, Random Forest (RF), Support Vector Machine (SVM) and Decision Tree-based AdaBoost (DT-AdaBoost), were carried out to construct the color quality evaluation model. Notably, the DT-Adaboost model shows significant potential for application in evaluating tea quality, with a correct discrimination rate (CDR) of 98.50% and a relative percent deviation (RPD) of 14.827 in the 266 samples used to verify the accuracy of the model. This result indicates that the integration of computer vision with machine learning models presents an effective approach for assessing the color quality of needle-shaped green tea.

Machine Learning Driven Atom-Thin Materials for Fragrance Sensing

Fragrance plays a crucial role in the daily lives. Its importance spans various sectors, from therapeutic purposes to personal care, making the understanding and accurate identification of fragrances essential. To fully harness the potential of fragrances, efficient and precise fragrance sensing and identification are necessary. However, current fragrance sensors face several limitations, particularly in detecting and differentiating complex scent profiles with high accuracy. To address these challenges, the use of atom-thin materials in fragrance sensors has emerged as a groundbreaking approach. These atom-thin sensors, characterized by their enhanced sensitivity and selectivity, offer significant improvements over traditional sensing technology. Moreover, the integration of Machine Learning (ML) into fragrance sensing has opened new opportunities in the field. ML algorithms applied to fragrance sensing facilitate advancements in four key domains: accurate fragrance identification, precise discrimination between different fragrances, improved detection thresholds for subtle scents, and prediction of fragrance properties. This comprehensive review delves into the synergistic use of atom-thin materials and ML in fragrance sensing, providing an in-depth analysis of how these technologies are revolutionizing the field, offering insights into their current applications and future potential in enhancing the understanding and utilization of fragrances.

Rapid and comprehensive grade evaluation of Keemun black tea using efficient multidimensional data fusion

To develop a comprehensive evaluation method for Keemun black tea, we used micro-near-infrared spectroscopy, computer vision, and colorimetric sensor array to collect data. We used support vector machine, least-squares support vector machine (LS-SVM), extreme learning machine, and partial least squares discriminant analysis algorithms to qualitatively discriminate between different grades of tea. Our results indicated that the LS-SVM model with mid-level data fusion attained an accuracy of 98.57% in the testing set. To quantitatively determine flavour substances in black tea, we used support vector regression. The correlation coefficient for the predicted sets of gallic acid, caffeine, epigallocatechin, catechin, epigallocatechin gallate, epicatechin, gallocatechin gallate and total catechins were 0.84089, 0.94249, 0.94050, 0.83820, 0.81111, 0.82670, 0.93230, and 0.93608, respectively. Furthermore, all compounds exhibited residual predictive deviation values exceeding 2. Hence, combining spectral, shape, colour, and aroma data with mid-level data can provide a rapid and comprehensive assessment of Keemun black tea quality.

A machine learning method for juice human sensory hedonic prediction using electronic sensory features

This study proposed a method that combines fused electronic sensory analysis technology with artificial neural network to predict the human sensory hedonic of fruit juice. Quantitative descriptive analysis (QDA) and the scoring test method were utilized for human sensory evaluation. The first step involved modeling the fused e-sensory features with human sensory attributes, followed by establishing a fitting model of human sensory attributes and acceptance. The R2 and RMSE values obtained were 0.77 and 0.42 (QDA method), and 0.63 and 0.63 (scoring test method). Finally, the relationship between the fusion e-sensory features and the human sensory hedonic was established. Model-1 achieved an R2 of 0.95 and an RMSE of 0.04, while model-2 achieved an R2 value of 0.88 and an RMSE value of 0.21. This study demonstrates the potential of fusing e-sensory technologies to replace human senses, which may lead to the development of devices with simultaneous multiple senses.

The impact of mulberry leaf extract at three different levels on reducing the glycemic index of white bread

In this study, the influences of mulberry leaf extract (MLE) addition on the physicochemical properties including the specific volume, texture and sensory features of white bread (WB) were evaluated by the sensory analysis technology. A double-blind, randomised, repeat-measure design was used to study the impact of MLE addition on the postprandial blood glucose response as well as the satiety index of WB. Results showed that the addition of MLE showed no significant effects on the physicochemical properties of WB except for the slight changes of color and bitterness. The addition of MLE significantly reduced the total blood glucose rise after ingestion of WB over 120 minutes, and reduced the GI value of WB in a dose-effect relationship. When the concentration of MLE reached 1.5 g per 100 g available carbohydrate, the GI value of WB could be reduced from 77 to 43. This study provides important information in terms of the appropriateness of MLE when added to more complex real food, the dose-dependent relationship could supply a reference for the application of MLE.

Optimization of the Mixed Gas Detection Method Based on Neural Network Algorithm

Real-time mixed gas detection has attracted significant interest for being a key factor for applications of the electronic nose (E-nose). However, mixed gas detection still faces the challenge of long detection time and a large amount of training data. Therefore, in this work, we propose a feasible way to realize low-cost fast detection of mixed gases, which uses only the part response data of the adsorption process as the training set. Our results indicated that the proposed method significantly reduced the number of training sets and the prediction time of mixed gas. Moreover, it can achieve new concentration prediction of mixed gas using only the response data of the first 10 s, and the training set proportion can reduce to 60%. In addition, the convolutional neural network model can realize both the smaller training set but also the higher accuracy of mixed gas. Our findings provide an effective way to improve the detection efficiency and accuracy of E-noses for the experimental measurement.

Flavor of tea (Camellia sinensis): A review on odorants and analytical techniques

Tea is among the most consumed nonalcoholic beverages worldwide. Understanding tea flavor, in terms of both sensory aspects and chemical properties, is essential for manufacturers and consumers to maintain high quality of tea products and to correctly distinguish acceptable or unacceptable products. This article gives a comprehensive review on the aroma and off-flavor characteristics associated with 184 odorants. Although many efforts have been made toward the characterization of flavor compounds in different types of tea, modern flavor analytical techniques that affect the results of flavor analysis have not been compared and summarized systematically up to now. Thus, the overview mainly provides the instrumental flavor analytical techniques for both aroma and taste of tea (i.e., extraction and enrichment, qualitative, quantitative, and chemometric approaches) as well as descriptive sensory analytical methodologies for tea, which is helpful for tea flavor researchers. Flavor developments of tea evolved toward time-saving, portability, real-time monitoring, and visualization are also prospected to get a deeper insight into the influences of different processing techniques on the formation and changes of flavor compounds, especially desired flavor compounds and off-flavor substances present at (ultra)trace amounts in tea and tea products.

Study on the Discrimination of Possible Error Sources That Might Affect the Quality of Volatile Organic Compounds Signature in Dairy Cattle Using an Electronic Nose

Simple Summary

In recent decades, remarkable progress in the development of electronic nose (EN) technologies, particularly for disease detection, has been accomplished through the disclosure of novel methods and associated devices, mainly for the detection of volatile organic compounds (VOCs). Herein, we assessed the ability of a novel EN technology (MENT-EGAS prototype) to respond to direct sampling and to evaluate the influence of possible error sources that might affect the quality of VOC signatures. Principal Component Analyses (PCA) evidenced the presence in the analyzed samples of sufficient information to consent the discrimination of different environmental backgrounds, feed headspaces and exhalated breath between two groups of cows fed with two different types of feed. Moreover, discrimination was also observed within the same group between exhalated breaths sampled before and after feed intake. Based on these findings, we provided evidence that the MENT-EGAS prototype can identify error sources with accuracy. Livestock precision farming technologies are powerful tools for monitoring animal health and welfare parameters in a continuous and automated way.

Abstract

Electronic nose devices (EN) have been developed for detecting volatile organic compounds (VOCs). This study aimed to assess the ability of the MENT-EGAS prototype-based EN to respond to direct sampling and to evaluate the influence of possible error sources that might affect the quality of VOC signatures. This study was performed on a dairy farm using 11 (n = 11) multiparous Holstein-Friesian cows. The cows were divided into two groups housed in two different barns: group I included six lactating cows fed with a lactating diet (LD), and group II included 5 non-lactating late pregnant cows fed with a far-off diet (FD). Each group was offered 250 g of their respective diet; 10 min later, exhalated breath was collected for VOC determination. After this sampling, 4 cows from each group were offered 250 g of pellet concentrates. Ten minutes later, the exhalated breath was collected once more. VOCs were also measured directly from the feed’s headspace, as well as from the environmental backgrounds of each. Principal component analyses (PCA) were performed and revealed clear discrimination between the two different environmental backgrounds, the two different feed headspaces, the exhalated breath of groups I and II cows, and the exhalated breath within the same group of cows before and after the feed intake. Based on these findings, we concluded that the MENT-EGAS prototype can recognize several error sources with accuracy, providing a novel EN technology that could be used in the future in precision livestock farming.

Electronic Sensor Technologies in Monitoring Quality of Tea: A Review

Tea, after water, is the most frequently consumed beverage in the world. The fermentation of tea leaves has a pivotal role in its quality and is usually monitored using the laboratory analytical instruments and olfactory perception of tea tasters. Developing electronic sensing platforms (ESPs), in terms of an electronic nose (e-nose), electronic tongue (e-tongue), and electronic eye (e-eye) equipped with progressive data processing algorithms, not only can accurately accelerate the consumer-based sensory quality assessment of tea, but also can define new standards for this bioactive product, to meet worldwide market demand. Using the complex data sets from electronic signals integrated with multivariate statistics can, thus, contribute to quality prediction and discrimination. The latest achievements and available solutions, to solve future problems and for easy and accurate real-time analysis of the sensory-chemical properties of tea and its products, are reviewed using bio-mimicking ESPs. These advanced sensing technologies, which measure the aroma, taste, and color profiles and input the data into mathematical classification algorithms, can discriminate different teas based on their price, geographical origins, harvest, fermentation, storage times, quality grades, and adulteration ratio. Although voltammetric and fluorescent sensor arrays are emerging for designing e-tongue systems, potentiometric electrodes are more often employed to monitor the taste profiles of tea. The use of a feature-level fusion strategy can significantly improve the efficiency and accuracy of prediction models, accompanied by the pattern recognition associations between the sensory properties and biochemical profiles of tea.

Moisture contents and product quality prediction of Pu-erh tea in sun-drying process with image information and environmental parameters

In this study, moisture contents and product quality of Pu-erh tea were predicted with deep learning-based methods. Images were captured continuously in the sun-drying process. Environmental parameters (EP) of air humidity, air temperature, global radiation, wind speed, and ultraviolet radiation were collected with a portable meteorological station. Sensory scores of aroma, flavor, liquor color, residue, and total scores were given by a trained panel. Convolutional neural network (CNN) and gated recurrent unit (GRU) models were constructed based on image information and EP, which were selected in advance using the neighborhood component analysis (NCA) algorithm. The evolved models based on deep-learning methods achieved satisfactory results, with RMSE of 0.4332, 0.2669, 0.7508 (also with R ² of .9997, .9882, .9986, with RPD of 53.5894, 13.1646, 26.3513) for moisture contents prediction in each batch of tea, tea at different sampling periods, the overall samples, respectively; and with RMSE of 0.291, 0.2815, 0.162, 0.1574, 0.3931 (also with R ² of .9688, .9772, .9752, .9741, .8906, with RPD of 5.6073, 6.5912, 6.352, 6.1428, 4.0045) for final quality prediction of aroma, flavor, liquor color, residue, total score, respectively. By analyzing and comparing the RMSE values, the most significant environmental parameters (EP) were selected. The proposed combinations of different EP can also provide a valuable reference in the development of a new sun-drying system.

Electrochemical Sensors and Biosensors for the Analysis of Tea Components: A Bibliometric Review

Tea is a popular beverage all around the world. Tea composition, quality monitoring, and tea identification have all been the subject of extensive research due to concerns about the nutritional value and safety of tea intake. In the last 2 decades, research into tea employing electrochemical biosensing technologies has received a lot of interest. Despite the fact that electrochemical biosensing is not yet the most widely utilized approach for tea analysis, it has emerged as a promising technology due to its high sensitivity, speed, and low cost. Through bibliometric analysis, we give a systematic survey of the literature on electrochemical analysis of tea from 1994 to 2021 in this study. Electrochemical analysis in the study of tea can be split into three distinct stages, according to the bibliometric analysis. After chromatographic separation of materials, electrochemical techniques were initially used only as a detection tool. Many key components of tea, including as tea polyphenols, gallic acid, caffeic acid, and others, have electrochemical activity, and their electrochemical behavior is being investigated. High-performance electrochemical sensors have steadily become a hot research issue as materials science, particularly nanomaterials, and has progressed. This review not only highlights these processes, but also analyzes and contrasts the relevant literature. This evaluation also provides future views in this area based on the bibliometric findings.

Recent Progress in Smart Electronic Nose Technologies Enabled with Machine Learning Methods

Machine learning methods enable the electronic nose (E-Nose) for precise odor identification with both qualitative and quantitative analysis. Advanced machine learning methods are crucial for the E-Nose to gain high performance and strengthen its capability in many applications, including robotics, food engineering, environment monitoring, and medical diagnosis. Recently, many machine learning techniques have been studied, developed, and integrated into feature extraction, modeling, and gas sensor drift compensation. The purpose of feature extraction is to keep robust pattern information in raw signals while removing redundancy and noise. With the extracted feature, a proper modeling method can effectively use the information for prediction. In addition, drift compensation is adopted to relieve the model accuracy degradation due to the gas sensor drifting. These recent advances have significantly promoted the prediction accuracy and stability of the E-Nose. This review is engaged to provide a summary of recent progress in advanced machine learning methods in E-Nose technologies and give an insight into new research directions in feature extraction, modeling, and sensor drift compensation.

Application of Multispectral Camera in Monitoring the Quality Parameters of Fresh Tea Leaves

The production of high-quality tea by Camellia sinensis (L.) O. Ktze is the goal pursued by both producers and consumers. Rapid, nondestructive, and low-cost monitoring methods for monitoring tea quality could improve the tea quality and the economic benefits associated with tea. This research explored the possibility of monitoring tea leaf quality from multi-spectral images. Threshold segmentation and manual sampling methods were used to eliminate the image background, after which the spectral features were constructed. Based on this, the texture features of the multi-spectral images of the tea canopy were extracted. Three machine learning methods, partial least squares regression, support vector machine regression, and random forest regression (RFR), were used to construct and train multiple monitoring models. Further, the four key quality parameters of tea polyphenols, total sugars, free amino acids, and caffeine content were estimated using these models. Finally, the effects of automatic and manual image background removal methods, different regression methods, and texture features on the model accuracies were compared. The results showed that the spectral characteristics of the canopy of fresh tea leaves were significantly correlated with the tea quality parameters (r ≥ 0.462). Among the sampling methods, the EXG_Ostu sampling method was best for prediction, whereas, among the models, RFR was the best fitted modeling algorithm for three of four quality parameters. The R2 and root-mean-square error values of the built model were 0.85 and 0.16, respectively. In addition, the texture features extracted from the canopy image improved the prediction accuracy of most models. This research confirms the modeling application of a combination of multi-spectral images and chemometrics, as a low-cost, fast, reliable, and nondestructive quality control method, which can effectively monitor the quality of fresh tea leaves. This provides a scientific reference for the research and development of portable tea quality monitoring equipment that has general applicability in the future.

Optimization of Electronic Nose Sensor Array for Tea Aroma Detecting Based on Correlation Coefficient and Cluster Analysis

The electronic nose system is widely used in tea aroma detecting, and the sensor array plays a fundamental role for obtaining good results. Here, a sensor array optimization (SAO) method based on correlation coefficient and cluster analysis (CA) is proposed. First, correlation coefficient and distinguishing performance value (DPV) are calculated to eliminate redundant sensors. Then, the sensor independence is obtained through cluster analysis and the number of sensors is confirmed. Finally, the optimized sensor array is constructed. According to the results of the proposed method, sensor array for green tea (LG), fried green tea (LF) and baked green tea (LB) are constructed, and validation experiments are carried out. The classification accuracy using methods of linear discriminant analysis (LDA) based on the average value (LDA-ave) combined with nearest-neighbor classifier (NNC) can almost reach 94.44~100%. When the proposed method is used to discriminate between various grades of West Lake Longjing tea, LF can show comparable performance to that of the German PEN2 electronic nose. The electronic nose SAO method proposed in this paper can effectively eliminate redundant sensors and improve the quality of original tea aroma data. With fewer sensors, the optimized sensor array contributes to the miniaturization and cost reduction of the electronic nose system.

Online Monitoring the Key Intermediates and Volatile Compounds Evolved from Green Tea Roasting by Synchrotron Radiation Photoionization Mass Spectrometry

Online monitoring of the volatile compounds during the tea roasting process is crucial to find the optimum roasting conditions and improve the quality of green tea. In this work, synchrotron radiation photoionization mass spectrometry (SR-PIMS) was utilized to online monitor the evolved gaseous compounds during the tea roasting process. By virtue of "soft" ionization and fast data acquisition characteristics of SR-PIMS, dozens of aroma compounds including alcohols, aldehydes, furans, and nitrogen- and sulfur-containing species were detected and identified in real time. Moreover, 5-hydroxymethylfurfural (5-HMF), the key intermediate of Maillard reactions, was found with high sensitivity. Evolution processes of all the products could be observed via the time- and temperature-resolved profiles in N₂ and the air. Dehydration was found to be the first step during roasting. Oxygen in the air was found to accelerate the formation rate of various stable species and intermediates in the course of the thermal treatment of fresh green tea. The formation mechanisms of evolved compounds such as three sulfur-containing compounds, i.e., dimethyl sulfide, hydrogen sulfide, and methanethiol, could be proposed according to the step-by-step formation process. The time-resolved results were demonstrated to be applicable in the evaluation of different roasting processes by statistical analysis. The optimum tea roasting temperature and duration are proposed to be around 200 °C and 1000 s.

Quality evaluation of Keemun black tea by fusing data obtained from near-infrared reflectance spectroscopy and computer vision sensors

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.

Rapid Identification of Different Grades of Huangshan Maofeng Tea Using Ultraviolet Spectrum and Color Difference

Tea is an important beverage in humans’ daily lives. For a long time, tea grade identification relied on sensory evaluation, which requires professional knowledge, so is difficult and troublesome for laypersons. Tea chemical component detection usually involves a series of procedures and multiple steps to obtain the final results. As such, a simple, rapid, and reliable method to judge the quality of tea is needed. Here, we propose a quick method that combines ultraviolet (UV) spectra and color difference to classify tea. The operations are simple and do not involve complex pretreatment. Each method requires only a few seconds for sample detection. In this study, famous Chinese green tea, Huangshan Maofeng, was selected. The traditional detection results of tea chemical components could not be used to directly determine tea grade. Then, digital instrument methods, UV spectrometry and colorimetry, were applied. The principal component analysis (PCA) plots of the single and combined signals of these two instruments showed that samples could be arranged according to grade. The combined signal PCA plot performed better with the sample grade descending in clockwise order. For grade prediction, the random forest (RF) model produced a better effect than the support vector machine (SVM) and the SVM + RF model. In the RF model, the training and testing accuracies of the combined signal were all 1. The grades of all samples were correctly predicted. From the above, the UV spectrum combined with color difference can be used to quickly and accurately classify the grade of Huangshan Maofeng tea. This method considerably increases the convenience of tea grade identification.

Emerging techniques for determining the quality and safety of tea products: A review

Spectroscopic techniques, electrochemical methods, nanozymes, computer vision, and modified chromatographic techniques are the emerging techniques for determining the quality and safety parameters (e.g., physical, chemical, microbiological, and classified parameters, as well as inorganic and organic contaminants) of tea products (such as fresh tea leaves, commercial tea, tea beverage, tea powder, and tea bakery products) effectively. By simplifying the sample preparation, speeding up the detection process, reducing the interference of other substances contained in the sample, and improving the sensitivity and accuracy of the current standard techniques, the abovementioned emerging techniques achieve rapid, cost-effective, and nondestructive or slightly destructive determination of tea products, with some of them providing real-time detection results. Applying these emerging techniques in the whole industry of tea product processing, right from the picking of fresh tea leaves, fermentation of tea leaves, to the sensory evaluation of commercial tea, as well as developing portable devices for real-time and on-site determination of classified and safety parameters (e.g., the geographical origin, grade, and content of contaminants) will not only eliminate the strong dependence on professionals but also help mechanize the production of tea products, which deserves further research. Conducting a review on the application of spectroscopic techniques, electrochemical methods, nanozymes, computer vision, and modifications of chromatographic techniques for quality and safety determination of tea products may serve as guide for other types of foods and beverages, offering potential techniques for their detection and evaluation, which would promote the development of the food industry.

An Optimum Tea Fermentation Detection Model Based on Deep Convolutional Neural Networks

Tea is one of the most popular beverages in the world, and its processing involves a number of steps which includes fermentation. Tea fermentation is the most important step in determining the quality of tea. Currently, optimum fermentation of tea is detected by tasters using any of the following methods: monitoring change in color of tea as fermentation progresses and tasting and smelling the tea as fermentation progresses. These manual methods are not accurate. Consequently, they lead to a compromise in the quality of tea. This study proposes a deep learning model dubbed TeaNet based on Convolution Neural Networks (CNN). The input data to TeaNet are images from the tea Fermentation and Labelme datasets. We compared the performance of TeaNet with other standard machine learning techniques: Random Forest (RF), K-Nearest Neighbor (KNN), Decision Tree (DT), Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), and Naive Bayes (NB). TeaNet was more superior in the classification tasks compared to the other machine learning techniques. However, we will confirm the stability of TeaNet in the classification tasks in our future studies when we deploy it in a tea factory in Kenya. The research also released a tea fermentation dataset that is available for use by the community.

Cross-Category Tea Polyphenols Evaluation Model Based on Feature Fusion of Electronic Nose and Hyperspectral Imagery

Tea polyphenols are important ingredients for evaluating tea quality. The rapid development of sensors provides an efficient method for nondestructive detection of tea polyphenols. Previous studies have shown that features obtained from single or multiple sensors yield better results in detecting interior tea quality. However, due to their lack of external features, it is difficult to meet the general evaluation model for the quality of the interior and exterior of tea. In addition, some features do not fully reflect the sensor signals of tea for several categories. Therefore, a feature fusion method based on time and frequency domains from electronic nose (E-nose) and hyperspectral imagery (HSI) is proposed to estimate the polyphenol content of tea for cross-category evaluation. The random forest and the gradient boosting decision tree (GBDT) are used to evaluate the feature importance to obtain the optimized features. Three models based on different features for cross-category tea (black tea, green tea, and yellow tea) were compared, including grid support vector regression (Grid-SVR), random forest (RF), and extreme gradient boosting (XGBoost). The results show that the accuracy of fusion features based on the time and frequency domain from the electronic nose and hyperspectral image system is higher than that of the features from single sensor. Whether based on all original features or optimized features, the performance of XGBoost is the best among the three regression algorithms (R² = 0.998, RMSE = 0.434). Results indicate that the proposed method in this study can improve the estimation accuracy of tea polyphenol content for cross-category evaluation, which provides a technical basis for predicting other components of tea.

Identification of Similar Chinese Congou Black Teas Using an Electronic Tongue Combined with Pattern Recognition

It is very difficult for humans to distinguish between two kinds of black tea obtained with similar processing technology. In this paper, an electronic tongue was used to discriminate samples of seven different grades of two types of Chinese Congou black tea. The type of black tea was identified by principal component analysis and discriminant analysis. The latter showed better results. The samples of the two types of black tea distributed on the two sides of the region graph were obtained from discriminant analysis, according to tea type. For grade discrimination, we determined grade prediction models for each tea type by partial least-squares analysis; the coefficients of determination of the prediction models were both above 0.95. Discriminant analysis separated each sample in region graph depending on its grade and displayed a classification accuracy of 98.20% by cross-validation. The back-propagation neural network showed that the grade prediction accuracy for all samples was 95.00%. Discriminant analysis could successfully distinguish tea types and grades. As a complement, the models of the biochemical components of tea and electronic tongue by support vector machine showed good prediction results. Therefore, the electronic tongue is a useful tool for Congou black tea classification.

Quality Evaluation of Green and Dark Tea Grade Using Electronic Nose and Multivariate Statistical Analysis

Aroma assessment remains difficult and uncertain in the present sensory assessment system. It is highly desirable to develop a new assessment method to discriminate the quality of various teas in the tea market. In the present work, based on linear discriminant analysis and principal component analysis, the aroma of dry and wet samples of different Xi-hu Longjing and Pu-erh teas were tested and differentiated by electronic noses (e-nose). The results confirm that e-nose can discriminate different priced Xi-hu Longjing tea samples in the range of 80-800 RMB/500 g and varying storage years of Pu-erh tea samples. Furthermore, for the detection of both dry and wet samples of Longjing and Pu-erh teas, the results reveal that all samples have specific aroma characteristics that e-nose can recognize. More importantly, contribution analysis in sensors indicates that nitrogen oxides, methane and alcohols are the characteristic components that contribute to the fragrances of different priced Xi-hu Longjing teas, while nitrogen oxides, aromatic benzene and amines make the fragrances of Pu-erh teas with different storage years disparate. PRACTICAL APPLICATION: This work demonstrates that e-nose can rapidly distinguish tea products with different price levels and varying storage years. With the advantages of ease of use, high portability and flexibility, e-nose will be widely expanded and applied in refined processing and the development of flavored foods.

The Electronic Nose Coupled with Chemometric Tools for Discriminating the Quality of Black Tea Samples In Situ

An electronic nose (E-nose), comprising eight metal oxide semiconductor (MOS) gas sensors, was used in situ for real-time classification of black tea according to its quality level. Principal component analysis (PCA) coupled with signal preprocessing techniques (i.e., time set value preprocessing, F1; area under curve preprocessing, F2; and maximum value preprocessing, F3), allowed grouping the samples from seven brands according to the quality level. The E-nose performance was further checked using multivariate supervised statistical methods, namely, the linear and quadratic discriminant analysis, support vector machine together with linear or radial kernels (SVM-linear and SVM-radial, respectively). For this purpose, the experimental dataset was split into two subsets, one used for model training and internal validation using a repeated K-fold cross-validation procedure (containing the samples collected during the first three days of tea production); and the other, for external validation purpose (i.e., test dataset, containing the samples collected during the 4th and 5th production days). The results pointed out that the E-nose-SVM-linear model together with the F3 signal preprocessing method was the most accurate, allowing 100% of correct predictive classifications (external-validation data subset) of the samples according to their quality levels. So, the E-nose-chemometric approach could be foreseen has a practical and feasible classification tool for assessing the black tea quality level, even when applied in-situ, at the harsh industrial environment, requiring a minimum and simple sample preparation. The proposed approach is a cost-effective and fast, green procedure that could be implemented in the near future by the tea industry.

The qualitative and quantitative assessment of tea quality based on E-nose, E-tongue and E-eye combined with chemometrics

Electronic nose (E-nose), electronic tongue (E-tongue) and electronic eye (E-eye) combined with chemometrics methods were applied for qualitative identification and quantitative prediction of tea quality. Main chemical components, such as amino acids, catechins, polyphenols and caffeine were measured by traditional methods. Feature-level fusion strategy for the integration of the signals was introduced to integrate the E-nose, E-tongue and E-eye signals, aiming at improving the performances of identification and prediction models. Perfect results with an accuracy of 100% were obtained for qualitative identification of tea quality grades, based on fusion signals by support vector machine and random forest. Quantitative models were established for predicting the contents of the chemical components based on independent electronic signals and fusion signals by partial least squares regression, support vector machine and random forest. Random forest based on the fusion signals achieved the best performance in predicting the concentration of those chemical components.

Odor Fingerprint Analysis Using Feature Mining Method Based on Olfactory Sensory Evaluation

In this paper, we aim to use odor fingerprint analysis to identify and detect various odors. We obtained the olfactory sensory evaluation of eight different brands of Chinese liquor by a lab-developed intelligent nose. From the respective combination of the time domain and frequency domain, we extract features to reflect the samples comprehensively. However, the extracted feature combined time domain and frequency domain will bring redundant information that affects performance. Therefore, we proposed data by Principal Component Analysis (PCA) and Variable Importance Projection (VIP) to delete redundant information to construct a more precise odor fingerprint. Then, Random Forest (RF) and Probabilistic Neural Network (PNN) were built based on the above. Results showed that the VIP-based models achieved better classification performance than PCA-based models. In addition, the peak performance (92.5%) of the VIP-RF model had a higher classification rate than the VIP-PNN model (90%). In conclusion, odor fingerprint analysis using a feature mining method based on the olfactory sensory evaluation can be applied to monitor product quality in the actual process of industrialization.

Development of an objective measure of quality and commercial value of Japanese-styled green tea (Camellia L. sinensis): the Quality Index Tool

A novel approach to evaluate the commercial value of green tea products is explored in this paper. The green tea Quality Index Tool (QI-Tool) is based on high performance liquid chromatography (HPLC), capable of identifying and understanding the constituents that are important to create superior consumer and commercially valuable green tea beverages in the Japanese-style. This tool will allow producers to better identify a product's potential value within the various levels of green tea retail quality structure. Via the quantification of theanine, caffeine and the catechins: epicatechin (EC), epicatechin gallate (ECG), epigallocatchin (EGC), epigallocatechin gallate (EGCG) and gallocatechin gallate (GCG) within a green tea beverage, the QI-Tool provides categorisation of a product against the green tea market retail competitive set. This allows a better understanding of the product's potential commercial value, as well as a comparison to other products within that market category. The QI-Tool is an alternative and promising method for objectively evaluating commercial value of green tea products using HPLC analysis.

Enhancing the Discrimination Ability of a Gas Sensor Array Based on a Novel Feature Selection and Fusion Framework

In this paper, a novel feature selection and fusion framework is proposed to enhance the discrimination ability of gas sensor arrays for odor identification. Firstly, we put forward an efficient feature selection method based on the separability and the dissimilarity to determine the feature selection order for each type of feature when increasing the dimension of selected feature subsets. Secondly, the K-nearest neighbor (KNN) classifier is applied to determine the dimensions of the optimal feature subsets for different types of features. Finally, in the process of establishing features fusion, we come up with a classification dominance feature fusion strategy which conducts an effective basic feature. Experimental results on two datasets show that the recognition rates of Database I and Database II achieve 97.5% and 80.11%, respectively, when k = 1 for KNN classifier and the distance metric is correlation distance (COR), which demonstrates the superiority of the proposed feature selection and fusion framework in representing signal features. The novel feature selection method proposed in this paper can effectively select feature subsets that are conducive to the classification, while the feature fusion framework can fuse various features which describe the different characteristics of sensor signals, for enhancing the discrimination ability of gas sensors and, to a certain extent, suppressing drift effect.

Mining Feature of Data Fusion in the Classification of Beer Flavor Information Using E-Tongue and E-Nose

Multi-sensor data fusion can provide more comprehensive and more accurate analysis results. However, it also brings some redundant information, which is an important issue with respect to finding a feature-mining method for intuitive and efficient analysis. This paper demonstrates a feature-mining method based on variable accumulation to find the best expression form and variables’ behavior affecting beer flavor. First, e-tongue and e-nose were used to gather the taste and olfactory information of beer, respectively. Second, principal component analysis (PCA), genetic algorithm-partial least squares (GA-PLS), and variable importance of projection (VIP) scores were applied to select feature variables of the original fusion set. Finally, the classification models based on support vector machine (SVM), random forests (RF), and extreme learning machine (ELM) were established to evaluate the efficiency of the feature-mining method. The result shows that the feature-mining method based on variable accumulation obtains the main feature affecting beer flavor information, and the best classification performance for the SVM, RF, and ELM models with 96.67%, 94.44%, and 98.33% prediction accuracy, respectively.

The Regular Interaction Pattern among Odorants of the Same Type and Its Application in Odor Intensity Assessment

The olfactory evaluation function (e.g., odor intensity rating) of e-nose is always one of the most challenging issues in researches about odor pollution monitoring. But odor is normally produced by a set of stimuli, and odor interactions among constituents significantly influenced their mixture’s odor intensity. This study investigated the odor interaction principle in odor mixtures of aldehydes and esters, respectively. Then, a modified vector model (MVM) was proposed and it successfully demonstrated the similarity of the odor interaction pattern among odorants of the same type. Based on the regular interaction pattern, unlike a determined empirical model only fit for a specific odor mixture in conventional approaches, the MVM distinctly simplified the odor intensity prediction of odor mixtures. Furthermore, the MVM also provided a way of directly converting constituents’ chemical concentrations to their mixture’s odor intensity. By combining the MVM with usual data-processing algorithm of e-nose, a new e-nose system was established for an odor intensity rating. Compared with instrumental analysis and human assessor, it exhibited accuracy well in both quantitative analysis (Pearson correlation coefficient was 0.999 for individual aldehydes (n = 12), 0.996 for their binary mixtures (n = 36) and 0.990 for their ternary mixtures (n = 60)) and odor intensity assessment (Pearson correlation coefficient was 0.980 for individual aldehydes (n = 15), 0.973 for their binary mixtures (n = 24), and 0.888 for their ternary mixtures (n = 25)). Thus, the observed regular interaction pattern is considered an important foundation for accelerating extensive application of olfactory evaluation in odor pollution monitoring.