Improved generalization of spectral models associated with Vis-NIR spectroscopy for determining the moisture content of different tea leaves

A review: Integration of NIRS and chemometric methods for tea quality control-principles, spectral preprocessing methods, machine learning algorithms, research progress, and future directions

With the steady rise in tea production, the need for effective tea quality monitoring has become increasingly pressing. Traditional sensory evaluation and wet chemical detection methods are insufficient for real-time tea quality monitoring. As an emerging technology, near infrared spectroscopy (NIRS) offers numerous advantages, such as preserving sample integrity, generating objective results, and enabling rapid, straightforward assessments. These features make it an ideal choice for real-time tea quality testing. This paper systematically reviews the principles of NIRS, spectral preprocessing methods, statistical modeling techniques, and commonly used machine learning approaches. Furthermore, it provides an in-depth discussion of the research progress of NIRS in areas such as fresh tea leaf quality evaluation, rapid detection of tea-specific components, tea quality assessment and species identification, geographic traceability, development of NIRS equipment, and standardization. Future research directions in the tea field are also proposed. This review serves as a valuable resource for researchers aiming to understand the application and development of NIRS technology in the tea field. It offers insights to facilitate real-time tea quality monitoring and ultimately achieve intelligent quality control.

Genome-Wide Identification, Characterization, and Expression Analysis of BES1 Family Genes in 'Tieguanyin' Tea Under Abiotic Stress

The BRI1-EMS-SUPPRESSOR 1 (BES1) family comprises plant-specific transcription factors, which are distinguished by atypical bHLH domains. Over the past two decades, genetic and biochemical studies have established that members of the BRI1-EMS-SUPPRESSOR 1 (BES1) family are crucial for regulating the expression of genes involved in brassinosteroid (BR) response in rapeseed. Due to the significance of the BES1 gene family, extensive research has been conducted to investigate its functional properties. This study presents a comprehensive identification and computational analysis of BES1 genes in 'Tieguanyin' (TGY) tea (Camellia sinensis). A total of 10 BES1 genes were initially identified in the TGY genome. Through phylogenetic tree analysis, this study uniquely revealed that CsBES1.2 and CsBES1.5 cluster with SlBES1.8 from Solanum lycopersicum, indicating their critical roles in fruit growth and development. Synteny analysis identified 20 syntenic genes, suggesting the conservation of their evolutionary functions. Analysis of the promoter regions revealed two types of light-responsive cis-elements, with CsBES1.4 exhibiting the highest number of light-related cis-elements (13), followed by CsBES1.9 and CsBES1.10. Additional validation via qRT-PCR experiments showed that CsBES1.9 and CsBES1.10 were significantly upregulated under light exposure, with CsBES1.10 reaching approximately six times the expression level of the control after 4 h. These results suggest that CsBES1.9 and CsBES1.4 could play crucial roles in responding to abiotic stress. This study offers novel insights into the functional roles of the BES1 gene family in 'Tieguanyin' tea and establishes a significant foundation for future research, especially in exploring the roles of these genes in response to abiotic stresses, such as light exposure.

A robust deep learning model for predicting green tea moisture content during fixation using near-infrared spectroscopy: Integration of multi-scale feature fusion and attention mechanisms

Fixation is a critical step in green tea processing, and the moisture content of the leaves after fixation is a key indicator of the fixation quality. Near-infrared spectroscopy (NIRS)-based moisture detection technology is often applied in the tea processing industry. However, temperature fluctuations during processing can cause changes in the NIRS curves, which in turn affect the accuracy of moisture prediction models based on the spectral data. To address this challenge, NIRS data were collected from samples at various stages of fixation and at different temperatures, and a novel deep learning network (DiSENet) was proposed, which integrates multi-scale feature fusion and attention mechanisms. Using a global modeling approach, the proposed method achieved a coefficient of determination (RP2) of 0.781 for moisture content prediction, with a root mean square error (RMSEP) of 1.720 % and a residual predictive deviation (RPD) of 2.148. On the dataset constructed for this study, DiSENet demonstrated superior predictive accuracy compared to the spectral correction methods of external parameter orthogonalization (EPO) and generalized least squares weighting (GLSW), as well as traditional global modeling methods such as partial least squares regression (PLSR) and support vector regression (SVR). This approach effectively corrects spectral interferences caused by temperature variations, thereby enhancing the accuracy of moisture content prediction. Thus, it offers a reliable solution for real-time, non-destructive moisture detection during tea processing.

Application of adaptive chaotic dung beetle optimization algorithm to near-infrared spectral model transfer

A new transfer approach was proposed to share calibration models of the hexamethylenetetramine-acetic acid solution for studying hexamethylenetetramine concentration values across different near-infrared (NIR) spectrometers. This approach combines Savitzky-Golay first derivative (S_G_1) and orthogonal signal correction (OSC) preprocessing, along with feature variable optimization using an adaptive chaotic dung beetle optimization (ACDBO) algorithm. The ACDBO algorithm employs tent chaotic mapping and a nonlinear decreasing strategy, enhancing the balance between global and local search capabilities and increasing population diversity to address limitations observed in traditional dung beetle optimization (DBO). Validated using the CEC-2017 benchmark functions, the ACDBO algorithm demonstrated superior convergence speed, accuracy, and stability. In the context of a partial least squares (PLS) regression model for transferring hexamethylenetetramine-acetic acid solutions using NIR spectroscopy, the ACDBO algorithm excelled over alternative methods such as uninformative variable elimination, competitive adaptive reweighted sampling, cuckoo search, grey wolf optimizer, differential evolution, and DBO in efficiency, accuracy of feature variable selection, and enhancement of model predictive performance. The algorithm attained outstanding metrics, including a determination coefficient for the calibration set (Rc2) of 0.99999, a root mean square error for the calibration set (RMSEC) of 0.00195%, a determination coefficient for the validation set (Rv2) of 0.99643, a root mean squared error for the validation set (RMSEV) of 0.03818%, residual predictive deviation (RPD) of 16.72574. Compared to existing OSC, slope and bias correction (S/B), direct standardization (DS), and piecewise direct standardization (PDS) model transfer methods, the novel strategy enhances the accuracy and robustness of model predictions. It eliminates irrelevant background information about the hexamethylenetetramine concentration, thereby minimizing the spectral discrepancies across different instruments. As a result, this approach yields a determination coefficient for the prediction set (Rp2) of 0.96228, a root mean squared error for the prediction set (RMSEP) of 0.12462%, and a relative error rate (RER) of 17.62331, respectively. These figures closely follow those obtained using DS and PDS, which recorded Rp2, RMSEP, and RER values of 0.97505, 0.10135%, 21.67030, and 0.98311, 0.08339%, 26.33552, respectively. Unlike conventional methods such as OSC, S/B, DS, and PDS, this novel approach does not require the analysis of identical samples across different instruments. This characteristic significantly broadens its applicability for model transfer, which is particularly beneficial for transferring specific measurement samples.

Application of visible near-infrared spectroscopy combined with colorimetric sensor array for the aroma quality evaluation in tencha drying process

The drying process is a critical stage in developing the aroma quality of tencha. In our research, visible near infrared (Vis-NIR) and colorimetric sensor array (Vis-NIR-CSA) were used for evaluating the aroma quality of tencha drying process. Vis-NIR recorded the spectral signal of CSA after the reaction in samples. Subsequently, the aroma quality was predicted by a combination of different data fusion strategies and classification and regression tree (CART) in tencha drying process. The high-level fusion strategy showed the best performance, with calibration and prediction set accuracy of 94.68% and 93.48%, respectively. The results indicated that Vis-NIR-CSA combined with high-level data fusion could be applied satisfactorily in the aroma quality evaluation of tencha. Moreover, pentanal was identified to be highly correlated with aroma quality during tencha drying process, which verified the sensor identification results. This study contributed to controlling good manufacturing practices and designing optimal tencha processing systems.

Effect of potato peel on the determination of soluble solid content by visible near-infrared spectroscopy and model optimization

The quantitative determination of the soluble solid content (SSC) of potatoes using NIR spectroscopy is useful for predicting the internal and external quality of potato products, especially fried products. In this study, the effect of peel on the partial least squares regression (PLSR) quantitative prediction of potato SSC was investigated by transmission and reflection. The results show that the variable sorting for normalization (VSN) pre-processing method improved model accuracy. Additive multiplicative scattering effects and intensity drift interference of the peels were reduced. The model accuracy reached a correlation coefficient of prediction (RP) of 0.85. The selection algorithm using variable combination population analysis and iterative retention of information variables (VCPA-IRIV) demonstrated that peel increases unnecessary information. When the effect of irrelevant variables was reduced, the results reached RP = 0.88 and the root mean square error of prediction (RMSEP) = 0.25 in the transmission mode was close to that of the full-wavelength peeled PLSR model (RP = 0.89 and RMSEP = 0.25). This indicates that the use of the combined algorithm (VSN-VCPA-IRIV) reduces the effect of the peel and enables samples with a peel to still be predicted accurately in the full-wavelength model. It also improves detection efficiency through the extraction of the necessary variables and optimizes the stability and accuracy of the model.

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.

Robustness and accuracy evaluation of moisture prediction model for black tea withering process using hyperspectral imaging

The rapid and non-destructive detection of moisture in withering leaves is an unsolved problem because the leaves are stacked together and have random orientation. To address this issue, this study aimed to establish more robust and accurate models. The performance of front side, back side and multi-region models were compared, and the front side model showed the worst transferability. Therefore, five effective wavelength (EW) selection algorithms were combined with a successive projection algorithm (SPA) to select EWs. It was found that the shuffled frog leaping algorithm (SFLA) combined with SPA was the best method for the front side model for moisture analyses. Based on the selected EWs, the extreme learning machine (ELM) became the model with the best self-verification result. Subsequently, moisture distribution maps of withering leaves were successfully generated. Considering the processing demand of withering leaves, local region models developed based on partial least squares and the SFLA-SPA method were applied to predict the moisture of withering leaves in the local and stacked region. The results showed that the RPD, R_cv and R_p values were above 1.6, 0.870 and 0.897, respectively. These results provide a useful reference for the non-destructive detection of moisture in withering leaves.

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

Tieguanyin is one of the top ten most popular teas and the representative of oolong tea in China. In this study, a rapid and non-destructive method is developed to detect adulterated tea and its degree. Benshan is used as the adulterated tea, which is about 0%, 10%, 20%, 30%, 40%, and 50% of the total weight of tea samples, mixed with Tieguanyin. Taking the fluorescence spectra from 475 to 1000 nm, we then established the 2-and 6-class discriminant models. The 2-class discriminant models had the best evaluation index when using SG-CARS-SVM, which can reach a 100.00% overall accuracy, 100.00% specificity, 100% sensitivity, and the least time was 1.2088 s, which can accurately identify pure and adulterated tea; among the 6-class discriminant models (0% (pure Tieguanyin), 10, 20, 30, 40, and 50%), with the increasing difficulty of adulteration, SNV-RF-SVM had the best evaluation index, the highest overall accuracy reached 94.27%, and the least time was 0.00698 s. In general, the results indicated that the two classification methods explored in this study can obtain the best effects. The fluorescence hyperspectral technology has a broad scope and feasibility in the non-destructive detection of adulterated tea and other fields.

Across different instruments about tobacco quantitative analysis model of NIR spectroscopy based on transfer learning

An instance transfer learning algorithm has been proposed based on weighted ELM to construct NIR quantitative analysis models across different instruments for tobacco.

Study on effects of airborne Pb pollution on quality indicators and accumulation in tea plants using Vis-NIR spectroscopy coupled with radial basis function neural network

Tea plants that have a large leaf area mainly suffer from heavy metal accumulation in the above-ground parts through foliar uptake. With the world rapid industrialization, this pollution in tea is considered a crucial challenge due to its potential health risks. The present study proposes an innovative approach based on visible and near-infrared (Vis-NIR) spectroscopy coupled with chemometrics for the characterization of tea chemical indicators under airborne lead stress, which can be performed fast and in situ. The effects of lead stress on chemical indicators and accumulation in leaves of the two tea varieties at different time intervals and levels of treatment were investigated. In addition, changes in cell structure and leaf stomata were monitored during foliar uptake of aerosol particles by transmission electron microscopy (TEM) and scanning electron microscopy (SEM). The spectral variation was able to classify the tea samples into the Pb treatment groups through the linear discriminant analysis (LDA) model. Two machine learning techniques, namely, partial least squares (PLS) and radial basis function neural network (RBFNN), were evaluated and compared for building the quantitative determination models. The RBFNN models combined with correlation-based feature selection (CFS) and PLS data compression methods were used to optimize the prediction performance. The results demonstrated that the PLS-RBFNN as a non-linear model outperformed the PLS model and provided the R-value of 0.944, 0.952, 0.881, 0.937, and 0.930 for prediction of MDA, starch, sucrose, fructose, glucose, respectively. It can be concluded that the proposed approach has strong application potential in monitoring the quality and safety of plants under airborne heavy metal stress.