Search

BACKGROUND: The acid value is an important indicator for evaluating the quality of edible oil during storage. This study employs a portable near‐infrared (NIR) spectroscopy system to determine the acid value during edible oil storage. Four MPA‐based variable selection methods, namely competitive adaptive reweighted sampling (CARS), the variable iterative space shrinkage approach (VISSA), iteratively variable subset optimization (IVSO), and bootstrapping soft shrinkage (BOSS) were introduced to optimize the preprocessed NIR spectra. Support vector machine (SVM) models based on characteristic spectra obtained by different selection methods were then established to achieve quantitative detection of the acid value during edible oil storage. RESULTS: The results revealed that, compared with the full‐spectrum SVM model, the SVM models established by the characteristic wavelengths optimized by the variable selection methods based on the MPA strategy exhibit a significant improvement in complexity and generalization performance. Furthermore, compared with the CARS, VISSA, and IVSO methods, the BOSS method obtained the least number of characteristic wavelength variables, and the SVM model established based on the optimized features of this method exhibited the optimal prediction performance. The root mean square error of prediction (RMSEP) was 0.11 mg g−1, the coefficient of determination (Rp2) was 0.92 and the ratio performance deviation (RPD) was 2.82, respectively. Conclusion: The overall results indicate that the variable selection methods based on the MPA strategy can select more targeted characteristic variables. This has good application prospects in NIR spectra feature optimization. © 2020 Society of Chemical Industry [ABSTRACT FROM AUTHOR]

Aims: Soluble solids content (SSC) is an essential indicator for evaluating the internal quality of fresh jujube, which can be used to classify the quality grade of fresh jujube. Methods: In this study, SSC was determined as the research index of the internal quality in Lingwu long jujube to classify their quality grade. The online rapid nondestructive grading device for jujube quality was designed, including the design of the hardware system and a software system. The performance of the device was evaluated by additional samples. Results: By grading external samples, the correct rate of classification of SSC was 86.7% (the first, second, and third grade were 100, 85.7, and 71.4%, respectively), and the residual predictive deviation (RPD) value of optimal model was 2.8 (>2.5). Conclusions: The acquired results revealed that, the device could be used in production. Practical Applications: In this study, we developed an online nondestructive sugar grading device for fresh jujubes, including the design of hardware systems, the development of software systems. NIR spectroscopy technique coupled with chemometric selection method of characteristic variables were used to build a prediction model for SSC in fresh jujubes. In addition, the model and device were evaluated by external samples, the accuracy of the result was high, and it could be used for the grading of the sugar content of fresh jujube and could potentially extend the quality parameter grading applied to other fruit. This study provided the basis for the development of nondestructive rapid grading system related to the quality of fruit and vegetable. [ABSTRACT FROM AUTHOR]

Apple is the most widely planted fruit in the world and is popular in consumers because of its rich nutritional value. In this study, the portable near‐infrared (NIR) transmittance spectroscopy coupled with temperature compensation and chemometric algorithms was applied to detect the storage quality of apples. The postharvest quality of apples including soluble solids content (SSC), vitamin C (VC), titratable acid (TA), and firmness was evaluated, and the portable spectrometer was used to obtain near‐infrared transmittance spectra of apples in the wavelength range of 590–1,200 nm. Mixed temperature compensation method (MTC) was used to reduce the influence of temperature on the models and to improve the adaptability of the models. Then, variable selection methods, such as uninformative variable elimination (UVE), competitive adaptive reweighted sampling (CARS), and successive projections algorithm (SPA), were developed to improve the performance of the models by determining characteristic variables and reducing redundancy. Comparing the full spectral models with the models established on variables selected by different variable selection methods, the CARS combined with partial least squares (PLS) showed the best performance with prediction correlation coefficient (Rp) and residual predictive deviation (RPD) values of 0.9236, 2.604 for SSC; 0.8684, 2.002 for TA; 0.8922, 2.087 for VC; and 0.8207, 1.992 for firmness, respectively. Results showed that NIR transmittance spectroscopy was feasible to detect postharvest quality of apples during storage. [ABSTRACT FROM AUTHOR]

BACKGROUND The study reports a portable near infrared (NIR) spectroscopy system coupled with chemometric algorithms for prediction of tea polyphenols and amino acids in order to index matcha tea quality. RESULTS: Spectral data were preprocessed by standard normal variate (SNV), mean center (MC) and first‐order derivative (1stD) tests. The data were then subjected to full spectral partial least squares (PLS) and four variable selection algorithms, such as random frog partial least square (RF‐PLS), synergy interval partial least square (Si‐PLS), genetic algorithm‐partial least square (GA‐PLS) and competitive adaptive reweighted sampling partial least square (CARS‐PLS). RF‐PLS was established and identified as the optimum model based on the values of the correlation coefficients of prediction (RP), root mean square error of prediction (RMSEP) and residual predictive deviation (RPD), which were 0.8625, 0.82% and 2.13, and 0.9662, 0.14% and 3.83, respectively, for tea polyphenols and amino acids. The content range of tea polyphenols and amino acids in matcha tea samples was 8.51–14.58% and 2.10–3.75%, respectively. The quality of matcha tea was successfully classified with an accuracy rate of 83.33% as qualified, unqualified and excellent grade. CONCLUSION: The proposed method can be used as a rapid, accurate and non‐destructive platform to classify various matcha tea samples based on the ratio of tea polyphenols to amino acids. © 2019 Society of Chemical Industry [ABSTRACT FROM AUTHOR]

Total fungi count (TFC) is a quality indicator of cocoa beans when unmonitored leads to quality and safety problems. Fourier transform near infrared spectroscopy (FT-NIRS) combined with chemometric algorithms like partial least square (PLS); synergy interval-PLS (Si-PLS); synergy interval-genetic algorithm-PLS (Si-GAPLS); Ant colony optimization – PLS (ACO-PLS) and competitive-adaptive reweighted sampling-PLS (CARS-PLS) was employed to predict TFC in cocoa beans neat solution. Model results were evaluated using the correlation coefficients of the prediction (Rp) and calibration (Rc); root mean square error of prediction (RMSEP), and the ratio of sample standard deviation to RMSEP (RPD). The developed models performance yielded 0.951 ≤ Rp ≤ 0.975; and 3.15 ≤ RPD ≤ 4.32. The models’ prediction stability improved in the order of PLS < CARS-PLS < ACO-PLS < Si-PLS < Si-GAPLS. FT-NIRS combined with Si-GAPLS may be employed for in-situ and noninvasive quantification of TFC in cocoa beans for quality and safety monitoring. [ABSTRACT FROM AUTHOR]

The degree of hydrolysis is one of the most important indexes for process control and quality assessment in proteins enzymatic hydrolysis. This article proposed a simple and rapid near infrared spectroscopy method for real-time quantifying the degree of hydrolysis in alcalase hydrolysis process. Efficient variables selection algorithms were systemically studied in multivariate calibrations; the partial least squares coupled with uninformative variables elimination and ant colony optimization were proposed for modeling with results yieldingRp= 0.9525. Additionally, 10 independent samples with the relative error less than 10% further confirmed the stability and reliability of this method. This work demonstrated that the near infrared spectroscopy technique with a selected multivariate calibration has a high potential forin situmonitoring of alcalase hydrolysis process in protein industry. [ABSTRACT FROM AUTHOR]

The acid value (AV) is an essential parameter for the quality and safety evaluation of peanut oil. In this study, for efficiently and real-time monitor of acid value (AV) in peanut oil, a portable spectroscopy system was first developed and combined with variables selection algorithms to measure acid value (AV) in peanut oils. Developed portable spectroscopy system was applied for transmittance spectrum data acquisition after which partial least squares (PLS) and several variables selection algorithms synergy interval partial least square (Si-PLS), genetic algorithm (GA), genetic algorithm combined with Si-PLS namely GA-Si-PLS, ant colony optimization (ACO) algorithms were systemically studied and comparatively used for modeling. The performances of these models were evaluated according to correlation coefficients squared in the prediction set ( R P ) and root mean square error of prediction (RMSEP). The results showed that the variables selection methods could select more significant variables and improve the model performance, especially for the GA-Si-PLS model with the best performance than other variables selection algorithms with R P = 0.9426 and RMSEP = 0.2980. Finally, the paper draws a conclusion that the developed portable spectroscopy system combined with a suitable variables selection methods could be used for the simultaneous and rapid measurement of acid value in peanut oil. [ABSTRACT FROM AUTHOR]

The approach presented herein reports the application of near infrared (NIR) spectroscopy, in contrast with human sensory panel, as a tool for estimating Chinese rice wine quality; concretely, to achieve the prediction of the overall sensory scores assigned by the trained sensory panel. Back propagation artificial neural network (BPANN) combined with adaptive boosting (AdaBoost) algorithm, namely BP-AdaBoost, as a novel nonlinear algorithm, was proposed in modeling. First, the optimal spectra intervals were selected by synergy interval partial least square (Si-PLS). Then, BP-AdaBoost model based on the optimal spectra intervals was established, called Si-BP-AdaBoost model. These models were optimized by cross validation, and the performance of each final model was evaluated according to correlation coefficient ( R p ) and root mean square error of prediction ( RMSEP ) in prediction set. Si-BP-AdaBoost showed excellent performance in comparison with other models. The best Si-BP-AdaBoost model was achieved with R p = 0.9180 and RMSEP = 2.23 in the prediction set. It was concluded that NIR spectroscopy combined with Si-BP-AdaBoost was an appropriate method for the prediction of the sensory quality in Chinese rice wine. [ABSTRACT FROM AUTHOR]

Amino acid nitrogen (AAN) is one of the most important indicators to assess the quality grade of soy sauce in China. Near infrared (NIR) spectroscopy technique combined with characteristic variable selection and extreme learning machine (ELM) was applied to detect AAN content in soy sauce in this work. First, the optimal spectral intervals were selected by synergy interval partial least square. Then, ELM model based on the optimal spectral intervals was established, called synergy interval extreme learning machine (Si-ELM) model. Support vector machine model based on the optimal intervals was established comparatively. These models were optimized by cross validation, and the performance of each final model was evaluated according to correlation coefficient ( $$ R_{\text{p}}^2 $$) and root mean square error of prediction (RMSEP) in prediction set. Si-ELM showed excellent performance. The best Si-ELM model was achieved with $$ R_{\text{p}}^2 = 0.9657 $$ and RMSEP = 0.0371 in the prediction set. It was concluded that NIR spectroscopy combined with Si-ELM was an appropriate method to detect AAN content in soy sauce. [ABSTRACT FROM AUTHOR]

Amino acid nitrogen and total acid are two most important quality indices to assess the quality of soy sauce in China. This work employed near infrared spectroscopy combined with synergy interval partial least square and genetic algorithm to detect amino acid nitrogen and total acid content in soy sauce. First, synergy interval partial least square was used to select efficient spectral regions from the full spectrum region; and then, genetic algorithm was used to selected variables from the efficient spectral regions, to build partial least square model. The optimal genetic algorithm synergy interval partial least square models were obtained as follows: Rc = 0.9988 and Rp = 0.9988 for amino acid nitrogen content model using 64 variables; Rc = 0.9917 and Rp = 0.9902 for total acid content model using 81 variables. Genetic algorithm synergy interval partial least square models showed superiority over the partial least square and synergy interval partial least square models. The results indicated that amino acid nitrogen and total acid content in soy sauce could be rapidly determined by near infrared spectroscopy technique. Also, the results indicated that genetic algorithm synergy interval partial least square can improve the performance in measurement of amino acid nitrogen and total acid content by near infrared spectroscopy. [ABSTRACT FROM AUTHOR]

Abstract: Total acid content (TAC) is an important index in assessing vinegar quality. This work attempted to determine TAC in vinegar using near infrared spectroscopy. We systematically studied variable selection and nonlinear regression in calibrating regression models. First, the efficient spectra intervals were selected by synergy interval PLS (Si-PLS); then, two nonlinear regression tools, which were extreme learning machine (ELM) and back propagation artificial neural network (BP-ANN), were attempted. Experiments showed that the model based on ELM and Si-PLS (Si-ELM) was superior to others, and the optimum results were achieved as follows: the root mean square error of prediction (RMSEP) was 0.2486g/100mL, and the correlation coefficient (R p) was 0.9712 in the prediction set. This work demonstrated that the TAC in vinegar could be rapidly measured by NIR spectroscopy and Si-ELM algorithm showed its superiority in model calibration. [Copyright &y& Elsevier]

Abstract: To rapidly and efficiently measure antioxidant activity (AA) in green tea, near infrared (NIR) spectroscopy was employed with the help of a regression tool in this work. Three different linear and nonlinear regressions tools (i.e. partial least squares (PLS), back propagation artificial neural network (BP-ANN), and support vector machine regression (SVMR)), were systemically studied and compared in developing the model. The model was optimized by a leave-one-out cross-validation, and its performance was tested according to root mean square error of prediction (RMSEP) and correlation coefficient (R p ) in the prediction set. Experimental results showed that the performance of SVMR model was superior to the others, and the optimum results of the SVMR model were achieved as follow: RMSEP =0.02161 and R p =0.9691 in the prediction set. The overall results sufficiently demonstrate that the spectroscopy coupled with the SVMR regression tool has the potential to measure AA in green tea. [Copyright &y& Elsevier]

Epigallocatechin-3-gallate (EGCG) is credited with the majority of the health benefits associated with green tea consumption. It has a high economic and medicinal value. The feasibility of using different variable selection approaches in Fourier transform near-infrared (FT-NIR) spectroscopy for a rapid and conclusive quantitative determination of EGCG in green tea was investigated. Graphically oriented multivariate calibration modeling procedures such as interval partial least squares (iPLS), synergy interval partial least squares (siPLS), and genetic algorithm optimization combined with siPLS (siPLS-GA) were applied to select the most efficient spectral variables that provided the lowest prediction error. The performance of the final model was evaluated according to the root mean square error of prediction (RMSEP) and coefficient of determination (R 2) for the prediction set. Experimental results showed that the siPLS-GA model obtained the best results in comparison to other models. The optimal models were achieved with R 2p = 0.97 and RMSEP=0.32. The model can be obtained with only 36 variables retained and it provides a robust model with good estimation accuracy. This demonstrates the potential of NIR spectroscopy with multivariate calibration methods to quickly detect the bioactive component in green tea. [ABSTRACT FROM AUTHOR]

Abstract: Total volatile basic nitrogen (TVB-N) content is one of important index of pork’s freshness, and Warner–Bratzler shear force (WBSF) is seen as the important index of pork’s tenderness. This paper attempted the feasibility to determine TVB-N content and WBSF in pork by Fourier transform near infrared (FT-NIR) spectroscopy. Synergy interval partial least square (SI-PLS) algorithm was performed to calibrate regression model. The number of PLS factors and the number of intervals were optimised simultaneously by cross-validation. The performance of the model was evaluated according to two correlation coefficients (R) in calibration and prediction sets. Experimental results showed that the correlations coefficients in the calibration set (Rc ) and prediction set (Rp ) were achieved as follows: Rc =0.8398 and Rp =0.8084 for TVB-N content model; Rc =0.7533 and Rp =0.7041 for WBSF model. The overall results demonstrated that NIR spectroscopy combined with SI-PLS could be utilised to determinate TVB-N content and WBSF in pork. [ABSTRACT FROM AUTHOR]

Abstract: NIR spectroscopy technique was attempted to measure total flavone content in snow lotus in this work. Interval partial least square with genetic algorithm (iPLS–GA) was used to select the efficient spectral regions and variables in model calibration. The performance of the final model was back-evaluated according to root mean square error of calibration (RMSEC) and correlation coefficient (R c ) in calibration set, and tested by mean square error of prediction (RMSEP) and correlation coefficient (R p ) in prediction set. The optimal iPLS–GA model was obtained with 6 PLS factors, when 5 spectral regions and 53 variables were selected. The measurement results of final model were achieved as follow: RMSEC (%)=0.8347/R c =0.9444 in the calibration set, and RMSEP (%)=1.0766/R p =0.9006 in the prediction set. Finally, iPLS–GA moded showed its excellent performance, when compared with other 5 different PLS models. This work demonstrated that total flavone content in snow lotus could be measured by NIR spectroscopy technique, and iPLS–GA revealed its superiority in model calibration. [Copyright &y& Elsevier]

Abstract: Near infrared (NIR) spectroscopy combined with multivariate calibration was attempted to analyze free amino acid content of Radix Pseudostellariae. The original spectra of Pseudostellariae samples in wavelength range of 10000–4000cm−1 were acquired. Partial least squares (PLS), kernel PLS (k-PLS), back propagation neural network (BP-NN), and support vector regression (SVR) algorithms were performed comparatively to develop calibration models. Some parameters of the calibration models were optimized by cross-validation. The performance of BP-NN model was better than PLS, k-PLS, and SVR models. The root mean square error of prediction (RMSEP) and the correlation coefficient (R) of BP-NN model were 0.687 and 0.889 in prediction set respectively. Results showed that NIR spectroscopy combined with multivariate calibration has significant potential in quantitative analysis of free amino acid content in Radix Pseudostellariae. [Copyright &y& Elsevier]

Abstract: Rapid discrimination of roast green tea according to geographical origin is crucial to quality control. Fourier transform near-infrared (FT-NIR) spectroscopy and supervised pattern recognition was attempted to discriminate Chinese green tea according to geographical origins (i.e. Anhui Province, Henan Province, Jiangsu Province, and Zhejiang Province) in this work. Four supervised pattern recognitions methods were used to construct the discrimination models based on principal component analysis (PCA), respectively. The number of principal components factors (PCs) and model parameters were optimized by cross-validation in the constructing model. The performances of four discrimination models were compared. Experimental results showed that the performance of SVM model is the best among four models. The optimal SVM model was achieved when 4 PCs were used, discrimination rates being all 100% in the training and prediction set. The overall results demonstrated that FT-NIR spectroscopy with supervised pattern recognition could be successfully applied to discriminate green tea according to geographical origins. [Copyright &y& Elsevier]

Abstract: This paper attempted the feasibility to determine content total polyphenols content in green tea with near infrared (NIR) spectroscopy coupled with an appropriate multivariate calibration method. Partial least squares (PLS), interval PLS (iPLS) and synergy interval PLS (siPLS) algorithms were performed comparatively to calibrate regression model. The number of PLS components and the number of intervals were optimized according to root mean square error of cross-validation (RMSECV) in calibration set. The performance of the final model was evaluated according to root mean square error of prediction (RMSEP) and correlation coefficient (R) in prediction set. Experimental results showed that the performance of siPLS model is the best in contrast to PLS and iPLS. The optimal model was achieved with R =0.9583 and RMSEP=0.7327 in prediction set. This study demonstrated that NIR spectroscopy with siPLS algorithm could be used successfully to analysis of total polyphenols content in green tea, and revealed superiority of siPLS algorithm in contrast with other multivariate calibration methods. [Copyright &y& Elsevier]

Abstract: Near-infrared (NIR) spectroscopy has been successfully utilized for the rapid identification of green, black and Oolong teas. The spectral features of each category are reasonably differentiated in the NIR region, and the spectral differences provided enough qualitative spectral information for identification. Support vector machine as a pattern recognition was applied to attain the differentiation of the three tea categories in this study. The top five latent variables are extracted by principal component analysis as the input of SVM classifiers. The identification results of the three tea categories were achieved by the RBF SVM classifiers and the polynomial SVM classifiers in different parameters. The best identification accuracies were up to 90%, 100% and 93.33%, respectively, when training, while, 90%, 100% and 95% when test. It was obtained using the RBF SVM classifier with σ =0.5. The overall results ensure that NIR spectroscopy combined with SVM discrimination method can be efficiently utilized for rapid and simple identification of the different tea categories. [Copyright &y& Elsevier]

Abstract: Near-infrared (NIR) spectroscopy has been successfully utilized for the rapid identification of green, black and Oolong tea. The spectral features of each tea category are reasonably differentiated in the NIR region, and the spectral differences provided enough qualitative spectral information for the identification of tea. Support vector machine (SVM) as the pattern recognition was applied to identify three tea categories in this study. The top five principal components (PCs) were extracted as the input of SVM classifiers by principal component analysis (PCA). The RBF SVM classifiers and the polynomial SVM classifiers were studied comparatively in this experiment. The best experimental results were obtained using the radial basis function (RBF) SVM classifier with σ =0.5. The accuracies of identification were all more than 90% for three tea categories. Finally, compared with the back propagation artificial neural network (BP-ANN) approach, SVM algorithm showed its excellent generalization for identification results. The overall results show that NIR spectroscopy combined with SVM can be efficiently utilized for rapid and simple identification of the tea categories. [Copyright &y& Elsevier]

Abstract: This paper indicates the possibility to use near infrared (NIR) spectroscopy as a rapid method to predict quantitatively the content of caffeine and total polyphenols in green tea. A partial least squares (PLS) algorithm is used to perform the calibration. To decide upon the number of PLS factors included in the PLS model, the model is chosen according to the lowest root mean square error of cross-validation (RMSECV) in training. The correlation coefficient R between the NIR predicted and the reference results for the test set is used as an evaluation parameter for the models. The result showed that the correlation coefficients of the prediction models were R =0.9688 for the caffeine and R =0.9299 for total polyphenols. The study demonstrates that NIR spectroscopy technology with multivariate calibration analysis can be successfully applied as a rapid method to determine the valid ingredients of tea to control industrial processes. [Copyright &y& Elsevier]

• The particle size and free amino acids of matcha were evaluated using Vis-NIR. • Different strategies for variable selection of Vis-NIR spectra were compared. • The ICPA-CARS-PLS evaluation models exhibited the optimum performance. Consumer preference for matcha is heavily influenced by its physicochemical properties. The visible-near infrared (Vis-NIR) spectroscopy technology coupled with multivariate analysis was investigated for rapid and non-invasive evaluation of particle size and the ratio of tea polyphenols to free amino acids (P/F ratio) of matcha. The multivariate selection algorithms such as synergy interval (Si), variable combination population analysis (VCPA), competitive adaptive reweighted sampling (CARS), and interval combination population analysis (ICPA) were compared, and eventually, the variable selection strategy of ICPA and CARS hybridization was firstly proposed for selecting the characteristic wavelengths from Vis-NIR spectra to build partial least squares (PLS) models. Results indicated that the ICPA-CARS-PLS models achieved satisfactory performance for the evaluation of matcha particle size (Rp = 0.9376) and P/F ratio (Rp = 0.9283). Hence the rapid, effectual, and nondestructive online monitoring, Vis-NIR reflectance spectroscopy in tandem with chemometric models is significant for the industrial production of matcha. [ABSTRACT FROM AUTHOR]

• Determination of procymidone residues in rapeseed oil by NIRS. • Feature optimization methods are compared for NIR spectral feature optimization. • Intelligent optimization algorithms are compared for SVM model parameter optimization. • This study verifies the usefulness of analysis of pesticide residues in edible oil by NIRS. The issue of pesticide residues has always been a hot topic at home and abroad. A method for the quantitative detection of procymidone residues in grain and oil products using near-infrared (NIR) spectroscopy has been proposed. First, a NIR spectrometer was used to collect spectral data from rapeseed oil samples with different concentrations of procymidone residues. Based on full-spectrum data, the wavelength points selected by bootstrapping soft shrinkage (BOSS) algorithm, competitive adaptive reweighted sampling (CARS) algorithm, and variable combination population analysis (VCPA) algorithm then were compared and were quantified using support vector regression (SVR) model. Simultaneously, the prediction results of the SVR model optimized by dung beetle optimizer (DBO) algorithm and pigeon-inspired optimization (PIO) algorithm were compared using the full-spectrum data. Finally, the wavelength selection algorithms and parameter optimization algorithms with the best prediction results were selected for comparison and combination. In light of the outcomes, the three spectral characteristic wavelength selection algorithms and the two optimization algorithms can improve the coefficient of determination (R P 2) and reduce the root mean square error of prediction (RMSEP). The SVR model that utilizing CARS and PIO algorithm demonstrates the best generalization performance among all models evaluated, and the R P 2 is 0.9939 with a RMSEP of 2.3435 mg·kg−1. The results indicate that the high-precision and rapid detection of procymidone in edible oil can be achieved using the SVR model optimized by input feature and parameter based on NIR spectral data. This has great significance in ensuring the safety of grain and oil food. [ABSTRACT FROM AUTHOR]

• A DTL method based on 1D-CNN and FT-NIR spectra to detect pesticide residues in corn oil is proposed. • The method enhances the accuracy of predicting chlorpyrifos residual levels in corn oil. • DTL addresses performance gaps between datasets and improves model generalization. • The developed DTL method provides an efficient and precise approach to detecting food safety issues. This study introduces a novel deep transfer learning (DTL) approach based on convolutional neural networks (CNN) and Fourier transform near-infrared (FT-NIR) spectroscopy to enhance the accuracy of predicting the residual levels of chlorpyrifos in corn oil. The method proposed involves creating a 1D-CNN model using existing data and utilizing DTL to enhance the performance of a new model by transferring the parameters learned from the trained model. The research findings demonstrate that compared to the CNN model, the proposed TL method achieves superior predictive accuracy while utilizing a smaller amount of FT-NIR spectral data, thus reducing the reliance on labeled FT-NIR spectral data for model training. Specifically, the coefficient of determination (R P 2) reaches 0.9754, and the relative percent deviation (RPD) is 6.4575. This study confirms that the developed DTL method based on CNN and FT-NIR provides an efficient and precise approach to detecting food safety issues. In addition, this method is not only applicable for pesticide residue detection in edible oils based on near-infrared spectroscopy but can also be used for the chemometric analysis of FT-NIR spectral data in other fields. [ABSTRACT FROM AUTHOR]

Instrumental test of black tea samples instead of human panel test is attracting massive attention recently. This study focused on an investigation of the feasibility for estimation of the color sensory quality of black tea samples using the VIS-NIR spectroscopy technique, comparing the performances of models based on the spectra and color information. In model calibration, the variables were first selected by genetic algorithm (GA); then the nonlinear back propagation-artificial neural network (BPANN) models were established based on the optimal variables. In comparison with the other models, GA-BPANN models from spectra data information showed the best performance, with the correlation coefficient of 0.8935, and the root mean square error of 0.392 in the prediction set. In addition, models based on the spectra information provided better performance than that based on the color parameters. Therefore, the VIS-NIR spectroscopy technique is a promising tool for rapid and accurate evaluation of the sensory quality of black tea samples. [ABSTRACT FROM AUTHOR]

• This study proposes a rapid detection method of fatty acid value of peanuts during storage. • Building a portable spectroscopy system to collect the spectra of peanut samples during storage. • Optimizing spectral characteristics using variable selection methods based on the MPA strategy. • Developing SVM models using the features optimized to determine fatty acid content of peanuts. In the process of peanut storage, the acidity index can be used as one of the important evaluation criteria of peanut storage characteristics. This study proposed a rapid detection method of acidity index of peanuts during storage based on a portable near-infrared (NIR) spectroscopy system. The portable spectroscopy system was built to collect the NIR spectra of peanuts during storage. The Savitzky-Golay (SG) smoothing combined with multiple scattering correction (MSC) was used to preprocess the raw spectra and normalize it. Three model population analysis-based (MPA-based) wavelength variable selection methods (namely: variable combination population analysis (VCPA), iterative retained information variable (IRIV) and (VCPA-IRIV) were employed comparatively to optimize the characteristics of the pre-treated spectra. And support vector machine (SVM) models were established based on optimized features to achieve rapid detection of acidity index of peanuts during storage. The results obtained showed that compared with the full-spectrum SVM model, the prediction performance of the SVM model based on optimized features has been improved. In addition, the VCPA-SVM model obtained the best prediction performance. The root mean square error (RMSEP) of the model was 0.61 g·kg−1, the coefficient of determination of the prediction set (R P 2) was 0.95, and the residual predictive deviation (RPD) was 4.31. The overall results demonstrate that the portable NIR spectroscopy system can realize the rapid detection of the acidity index during peanut storage, and the VCPA algorithm can efficiently obtain the characteristic wavelengths of the NIR spectra of peanuts during storage. [ABSTRACT FROM AUTHOR]

The non-sugar solids (NSS) content is one of the most important nutrition indicators of Chinese rice wine. This study proposed a rapid method for the measurement of NSS content in Chinese rice wine using near infrared (NIR) spectroscopy. We also systemically studied the efficient spectral variables selection algorithms that have to go through modeling. A new algorithm of synergy interval partial least square with competitive adaptive reweighted sampling (Si-CARS-PLS) was proposed for modeling. The performance of the final model was back-evaluated using root mean square error of calibration ( RMSEC ) and correlation coefficient ( R c ) in calibration set and similarly tested by mean square error of prediction ( RMSEP ) and correlation coefficient ( R p ) in prediction set. The optimum model by Si-CARS-PLS algorithm was achieved when 7 PLS factors and 18 variables were included, and the results were as follows: R c = 0.95 and RMSEC = 1.12 in the calibration set, R p = 0.95 and RMSEP = 1.22 in the prediction set. In addition, Si-CARS-PLS algorithm showed its superiority when compared with the commonly used algorithms in multivariate calibration. This work demonstrated that NIR spectroscopy technique combined with a suitable multivariate calibration algorithm has a high potential in rapid measurement of NSS content in Chinese rice wine. [ABSTRACT FROM AUTHOR]

• This study compares the potential of using different NIR spectra to detect acidity index of peanut. • Selecting spectral intervals of FT-NIR and P-NIR spectra by the SiPLS algorithm. • Optimizing characteristic wavelength variables using the BOSS algorithm. • Developing SVM models using the features optimized to determine acidity index of peanuts. This study compared the potential of detection models based on different near-infrared (NIR) spectral characteristics in detecting acidity index of peanut during storage. Fourier transform near-infrared (FT-NIR) spectrometer and portable near-infrared (P-NIR) spectrometer were used to obtain the NIR spectra of peanut samples in different storage periods. The characteristic wavelength intervals of the preprocessed NIR spectra were roughly optimized by synergy interval partial least squares (SiPLS). The bootstrapping soft shrinkage (BOSS) was introduced to further fine select the characteristic wavelength variables, and the support vector machine (SVM) models based on the optimized characteristic wavelength variables were established. The results obtained showed that the SVM model based on the fusion of different NIR spectra wavelength variables obtained the best predictive performance. Root mean square error of prediction set (RMSEP) of the model was 0.73 g·kg−1, the determination coefficient of prediction set (R P 2) was 0.93, and the residual prediction deviation (RPD) was 3.83. The overall results indicate that although the commercial NIR spectrometer and portable near-infrared spectroscopy system overlap in band, the wavelength variables obtained can play a complementary effect to a certain extent. Therefore, the combination of two instrument data can effectively improve the generalization performance of the detection model. [ABSTRACT FROM AUTHOR]

Total volatile basic nitrogen (TVB-N) is one of the key indicators for assessing fish freshness. This research employed near-infrared (NIR) and Raman spectroscopy methods to detect the TVB-N content in snakehead fillets. We extracted feature variables associated with TVB-N from NIR and Raman spectroscopy using Variable Crossover Point Arithmetic - Improved Reduced-Input Vector (VCPA-IRIV). Using these features, we established partial least squares (PLS) and One-dimensional Convolutional Neural Network (1D-CNN) models. Subsequently, data fusion strategies were employed to predict the TVB-N content. Notably, feature-level fusion in conjunction with Bayesian-optimized 1D-CNN, reached the best results, as evidenced by calibration and predictive correlation coefficients of 0.9677 and 0.9676 for TVB-N. These findings underscore the effectiveness of both NIR and Raman spectroscopy in evaluating fish freshness. The fusion of these two vibrational spectroscopy techniques enables a more rapid, efficient and comprehensive quantification of fish freshness. [Display omitted] • NIR and Raman spectroscopy were used to predict TVB-N in snakehead. • Both NIR and Raman produced well in evaluating TVB-N. • Bayesian-optimized 1D-CNN improved model performance to predict TVB-N. • The fusion of NIR and Raman spectral achieved optimum prediction results. [ABSTRACT FROM AUTHOR]

The meat processing industry faces challenges in maintaining the gel quality in minced chicken products, which affects consumer appeal and overall product quality. This study investigates the use of ultrasonic treatment to improve the gel quality of minced chicken and employs Near-Infrared (NIR) and Raman spectroscopy for rapid, non-destructive gel strength assessment. Initially, ultrasonic treatment was applied at various durations, with optimal results observed at approximately 30 min, significantly improving gel strength, texture profile, and reducing centrifugal loss rate. To comprehensively assess gel strength, Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) models were established using individual NIR and Raman spectral data, as well as their fusion. The LSTM model with fused NIR-Raman data demonstrated superior performance (Rp2 = 0.9882, RPD = 9.2091), outperforming individual techniques and CNN-based models. This study demonstrates that ultrasonic treatment can effectively improve minced chicken gel quality, while the fusion of NIR and Raman spectroscopy coupled with LSTM deep learning offers a reliable, non-destructive, and rapid method for predicting gel strength. This approach addresses the industry's need for innovative quality assessment methods, potentially improving product quality and consumer satisfaction in the processed chicken market. • Ultrasonic treatment improves minced chicken gel quality. • Fusing of NIR and Raman spectroscopy was proposed to comprehensively assess the gel quality change. • Spectral data fusion strategy improves predictive model performance. • Deep learning algorithms applied to spectral data for accurate quality assessment. • LSTM with fused NIR-Raman data outperforms other models in gel strength prediction. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) technology is advancing the digitization and intelligence development of the food industry. A promising application is using deep learning-assisted visible near-infrared (vis-NIR) spectroscopy to monitor residual sugar and bacterial concentration in real-time, ensuring kombucha quality during production. The feature fingerprints of residual sugar and bacterial concentration were extracted by four variable selection algorithms and then reconstructed using serial and parallel processing methods. Based on these reconstructed features, Partial Least Squares (PLS) and Convolutional Neural Networks (1DCNN and 2DCNN) models were developed and compared. The experimental results showed that the 2DCNN model based on reconstruction features achieved superior performance. The RPDs of the residual sugar and bacterial concentrations models were 4.49 and 6.88, while the MAEs were 0.42 mg/mL and 0.04 (Abs), respectively. These results suggest that the proposed modeling strategy effectively supports quality control during kombucha production and provides a new perspective for spectral analysis. • An on-line visible-near infrared spectroscopy monitoring system was developed. • 4 variable selection algorithms were used to extract fingerprint features. • The feature fingerprint was reconstructed by serial and parallel methods. • Reconstruction features were processed using image and data enhancement. • Convolutional neural network was constructed based on reconstructed features. [ABSTRACT FROM AUTHOR]

The quality of edible oils is closely related to their chemical compositions. Antioxidants have widespread application in edible oil production. In this study, a pioneering detection approach involving the use of a one-dimensional convolutional autoencoder (1D-CAE) was introduced to compress spectral data for assessing antioxidant levels in edible oils. Fourier-transform near-infrared (FT-NIR) characterisation of edible oil samples with varying antioxidant concentrations was also conducted. An 1D-CAE model was developed to compress different pre-processed spectra into a condensed representation. These compressed features were then integrated with a support vector machine and partial least squares regression models to establish correlations for each target. The study examined the influence of pre-processing steps and feature engineering methods on near-infrared spectral analysis through independent or combined model analysis. The findings revealed that features derived from the 1D-CAE model demonstrated remarkable repeatability and can be utilised to construct robust detection models. The experimental results showed that the optimal detection model derived based on the 1D-CAE compression features has an average R2, RPD and RMSE of 0.9953, 15.1664 and 1.2035, respectively, on the prediction set. FT-NIR spectroscopy can be used to accurately detect butylated hydroxytoluene in edible oils. Therefore, autoencoders are an effective tool in spectroscopic analysis, offering promising avenues for future research and application. • Unsupervised autoencoders (AE) facilitated the compression of spectral features. • Preprocessing and feature engineering could be circumvented in the spectrum analysis. • The features extracted by the AE could be utilised for the detection of BHT in edible oils. [ABSTRACT FROM AUTHOR]

Highlights: [•] Nondestructive measurement of TVB-N in pork meat by integrating three techniques. [•] Correlation analysis of the three sensors data with TVB-N content of pork meat. [•] Extraction of the optimum feature variables from three sensors data. [•] Data fusion based on feature variables and BP-ANN model for measuring TVB-N content. [ABSTRACT FROM AUTHOR]

A neoteric measure for quantitative assay of Aflatoxin B 1 (AFB 1) in maize based on an optimized feature model of near-infrared (NIR) spectroscopy was proposed in the work. A portable near-infrared spectroscopy system constructed by the group was employed to collect maize samples with varying degrees of mildew. The variable selection methods of interval variable iterative space shrinkage approach (IVISSA), iterative retained information variable (IRIV), and particle swarm optimization combined moving window (PSO-CMW) were introduced to perform feature selection on the pretreatment NIR spectra. The characteristic wavelength variables after screening were used to constitute support vector machine (SVM) and partial least squares (PLS) test model respectively to implement the measurement of AFB 1 in maize, and the detection performance of the two types of models was compared. The results obtained showed that the overall performances of SVM models were higher than that of PLS models, and the SVM model based on the characteristic wavelength variables optimized by the PSO-CMW method had the most prominent generalization performance. The root mean square error of prediction (RMSEP) of the model was 3.5967 μg kg−1, the coefficient of determination (R P 2) was 0.9707, and the relative prediction deviation (RPD) was 5.7538. The overall results demonstrate that the optimized features of NIR spectra can realize the on-site quick testing of the AFB 1 in maize with high precision by constructing a nonlinear SVM detection model. This investigation provides an original approach for speedy quantitative detection of mycotoxins in cereals. [Display omitted] • This study proposes a neoteric measure for the quantitative assay of AFB 1 in maize. • A portable spectroscopy system was built for spectral data acquisition. • Optimizing characteristic wavelength variables using different chemometrics methods. • Developing SVM models using the features optimized to determine the AFB 1 in maize. [ABSTRACT FROM AUTHOR]

Abstract: Near infrared (NIR) spectroscopy combined with support vector data description (SVDD) was attempted to identify geographical origins of Radix Pseudostellariae. Original spectra of eggs in wavelength range of 10000–4000cm−1 were acquired. SVDD was performed to calibrate discrimination model, and some parameters of SVDD model were optimized. Meanwhile, conversional two-class classification method—support vector machine (SVM) was used comparatively for classification. Compared with SVM classification, SVDD model showed its superior ability in dealing with imbalance training samples. When the proportion of the number of Radix Pseudostellariae from Anhui province (the area where genuine crude Radix Pseudostellariae was cultivated) and other provinces was one to sixteen, the identification rate of SVDD model was 92.5% in prediction set. This work indicates that NIR spectroscopy combined with SVDD is an excellent choice in building one-class calibration model for discrimination of genuine crude Radix Pseudostellariae. [Copyright &y& Elsevier]

Abstract: Near infrared (NIR) spectroscopy combined with pattern recognition was attempted to discriminate egg’s freshness. A new algorithm support vector data description (SVDD) was employed to solve the classification problem due to imbalance number of training samples. Original spectra of eggs in wavelength range of 10,000–4000cm−1 were acquired. SVDD was performed to calibrate discrimination model, and some parameters of SVDD model were optimized. Meanwhile, several conversional two-class classification methods (i.e. partial least square discrimination analysis, PLS-DA; K-nearest neighbors, KNN; artificial neural network, ANN; support vector machine, SVM) were also used comparatively for classification. Experimental results showed that SVDD got better performance than the conversional classification models in same condition. The identification rates of fresh eggs and unfresh eggs were both 93.3%. This work indicates that it is feasible to identify egg’s freshness using NIR spectroscopy, and SVDD is an excellent choice in solving the problem of imbalance number of training samples. [Copyright &y& Elsevier]

This study built a portable NIR spectroscopy system to acquire the NIR spectra of wheat samples at disparate storage stages by means of diffuse reflectance, and the obtained NIR spectra were pre-processed appropriately. In order to obtain highly targeted feature wavelength variables, three variable selection methods were used to optimize the feature wavelength of the pre-processed NIR spectra. Finally, PLS quantitative detection models were developed on the basis of optimized characteristic wavelength variables to realize rapid detection of the AFB1 in wheat during storage, and the outcomes of each PLS model were compared and analyzed. The figure shows the connection between the modules and the near-infrared spectroscopy collection process of wheat samples. [Display omitted] • A portable NIRS system was developed to determine the AFB 1 in wheat during storage. • The characteristic wavelengths were optimized by different variable selection algorithm. • PLS models were established using selected features to determine the AFB 1 in wheat. • This study provides theoretical reference for more targeted spectrometer development. Wheat is a widely grown grain crop around the world and is highly susceptible to environmental factors during storage and transportation, resulting in the production of fungal toxins that are harmful to humans. Of these, aflatoxin B 1 (AFB 1) is the most prevalent and most toxic. In view of this, this study used a self-built portable near-infrared spectroscopy system to predict the AFB 1 content of wheat during storage and investigated and compared the prediction effects of different wavelength selection algorithms on the constructed PLS model. Firstly, the NIR spectra of wheat samples at disparate storage stages were acquired using the NIR spectroscopy system. Secondly, the raw NIR spectra were pretreated by Savizkg-Golag (SG) smoothing, standard normal variate (SNV) and normalization in turn. Finally, three variable optimization methods, which were variable combination population analysis (VCPA), variable iterative space shrinkage approach (VISSA) and competitive adaptive reweighted sampling (CARS), were applied to select the characteristic wavelength variables of the pre-processed spectra. Partial least squares (PLS) models based on the optimized features of the three methods were established, respectively. The results obtained showed that the CARS-PLS model had the best overall effect. The root mean square error of prediction (RMSEP) for the best CARS-PLS was 2.0965 μ g ∙ k g -1, the prediction coefficient of determination (R p 2) was 0.9935, and the ratio of prediction to deviation (RPD) was 7.3279. The CARS variable screening method was used to effectively select the characteristic wavebands associated with AFB 1 in wheat, compressing the number of wavelength variables, simplifying the model structure and improving model performance. The results reveal that the self-built portable NIR spectroscopy system enables to determine the AFB 1 in wheat during storage. Furthermore, through the feature optimization of spectral wavelength variables can effectively exclude undesired wavelength variable information. [ABSTRACT FROM AUTHOR]

The fatty acid value of rice is one of the important indexes to reflect its freshness. A portable near-infrared spectroscopy (NIRS) system was designed and assembled to dynamically monitor fatty acid values in rice storage in this study. First, the near-infrared (NIR) spectra of rice samples in different storage periods were obtained using the portable NIRS system. Then, a weighted multiplicative scatter correction with variable selection (WMSCVS) algorithm was applied to the original spectra for scattering correction, and to compress variable space for achieving characteristic wavelengths. Finally, a partial least square (PLS) regression model was established using the characteristic wavelengths to realize the rapid monitoring of fatty acid values in rice storage. The results showed that the performance of the optimal PLS model based on the features by the WMSCVS algorithm was significantly better than those of the optimal PLS models based on SNV and MSC pre-processing spectra, with the determination coefficient (R P 2) of 0.9615 and the root mean square error of prediction (RMSEP) of 0.3626 in the predictive process. The overall results demonstrate that it is feasible to use the portable NIRS system developed by our team to quickly monitor the fatty acid values in rice storage. • Fatty acid value can be used as one of the criteria for determining the rice quality. • A portable NIRS system was developed to dynamically monitor fatty acid values in rice storage. • The WMSCVS algorithm was applied to correct and optimize original NIR spectra. • Developing a PLS model using optimized wavelengths to monitor the fatty acid value of rice. [ABSTRACT FROM AUTHOR]

• Fatty acid value is one of the important indexes to judge wheat flour quality during storage. • Fatty acid value was detected by a self-built portable near infrared spectrometer. • The characteristic wavelengths were optimized by the VCPA algorithm. • Developing ELM models using selected feature for detection of fatty acid value. Fatty acid value is one of the important indexes to judge wheat flour quality during storage. A portable near-infrared (NIR) spectroscopy system was developed established for the quantitative detection of fatty acids in wheat flour during storage. First, the portable NIR spectroscopy system was used to obtain the spectra of wheat flour in different storage periods, and the spectra acquired were corrected by standard normal variate (SNV) method. Then, variable combination population analysis (VCPA) was used to optimize the characteristic wavelength variables of the SNV corrected spectra, and the characteristic wavelength variables highly related to the fatty acid value were determined. Finally, extreme learning machine (ELM) was employed to construct quantitative detection models based on different characteristic wavelength variables to achieve quantitative detection of fatty acid value. In the process, the effects of the "Sigmoidal" and "Sine" activation functions on the performance of the ELM model were compared. The experimental results showed that in this study, the two activation functions have little effect on the generalization performance of the ELM model. The ELM models based on different input characteristic wavelength variables all showed good prediction accuracy and stability when predicting independent samples in the validation set, and the mean of R P 2 from the ELM model in each mode was above 0.96. The overall results demonstrate that it is feasible to use the portable NIR spectroscopy system built combined with appropriate chemometric methods to achieve quantitative determination of fatty acid values in wheat flour during storage. In addition, the VCPA algorithm has a good application prospect in the optimization of NIR spectral characteristic wavelengths. [ABSTRACT FROM AUTHOR]

[Display omitted] • A self-built olfactory sensor including three color sensitive dyes were developed. • Visible near infrared was used to expand the dimension of aroma information. • Different levels of data fusion strategies were implemented. • The aroma quality during tencha drying process was accurately predicted. • Volatile compounds differences between tencha drying process were verified. 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. [ABSTRACT FROM AUTHOR]

[Display omitted] • A rapid and efficient technology for on-line monitoring kombucha is developed. • Near-infrared spectroscopy is used as a monitoring tool for total sugars in kombucha. • Linear and nonlinear calibration methods are combined and compared. • The combined model can be used to monitor the total sugar. Kombucha is widely recognized for its health benefits, and it facilitates high-quality transformation and utilization of tea during the fermentation process. Implementing on-line monitoring for the kombucha production process is crucial to promote the valuable utilization of low-quality tea residue. Near-infrared (NIR) spectroscopy, together with partial least squares (PLS), backpropagation neural network (BPANN), and their combination (PLS-BPANN), were utilized in this study to monitor the total sugar of kombucha. In all, 16 mathematical models were constructed and assessed. The results demonstrate that the PLS-BPANN model is superior to all others, with a determination coefficient (R2p) of 0.9437 and a root mean square error of prediction (RMSEP) of 0.8600 g/L and a good verification effect. The results suggest that NIR coupled with PLS-BPANN can be used as a non-destructive and on-line technique to monitor total sugar changes. [ABSTRACT FROM AUTHOR]

[Display omitted] • For detection of fatty acids, NIR spectroscopy in combination with CSA has been used. • Low and mid-level data fusion were used to build a stable model. • The best intervals from mobile NIR spectral data were selected using Si-PLS. • Mid-level fusion yielded better results for the prediction of fatty acids. The use of sensor fusion, a novel method of combining artificial senses, has become increasingly popular in the assessment of food quality. This study employed a combination of the colorimetric sensor array (CSA) and mobile near-infrared (NIR) spectroscopy to predict free fatty acids in wheat flour. In conjunction with a partial least squares model, Low- and mid-level fusion strategies were used for quantification. Accordingly, performance of the built model was evaluated based on higher correlation coefficients between calibration and prediction (R C and R P), lower root mean square error of prediction (RMSEP), and a higher residual predictive deviation (RPD). The mid-level fusion coupled PLS model produced superior data fusion findings, with R C = 0.8793, RMSECV = 7.91 mg/100 g, R P = 0.8747, RMSEP = 6.99 mg/100 g, and RPD = 2.27. The findings of the study suggest that the NIR-CSA fusion approach could be effectively applied to the prediction of free fatty acids in wheat flour. [ABSTRACT FROM AUTHOR]

Phytochemically, peanuts provide significant nutritional value to humans, animals, and the food industry as a whole. This study was conducted to explore the potential for near-infrared (NIR) spectroscopy and effective variable selection algorithms to quantify phenolic compounds in peanut seeds. The phenolics were extracted and then identified and quantified using high-performance liquid chromatography (HPLC). The spectroscopic data were acquired from the peanut samples using a tabletop NIR spectrometer with a wavelength range of 10,000–4000 cm–1. The acquired spectra were preprocessed using a synergistic effect of first- and second-order derivative (FOD and SOD) preprocessing techniques, and multivariate algorithms were used, examined, and evaluated using correlation coefficients of the validation set (R p), root mean square error of prediction (RMSEP), and residual predictive deviations (RPDs). The competitive adaptive reweighted sampling-partial least squares (CARS-PLS) model produced optimum performance for chlorogenic acid (R p = 0.933, RPD = 2.77), kaempferol (R p = 0.928, RPD = 2.68), p -coumaric acid (R p = 0.900, RPD = 2.32), and quercetin (R p = 0.932, RPD = 2.88), respectively. Therefore, this study proved that NIR spectroscopy in combination with CARS-PLS was capable of nondestructively predicting phenolic content in peanut seeds. • Phenolic compounds in peanut seeds were individually quantified using HPLC. • NIR spectroscopy was effectively employed to evaluate phenolic compounds. • Spectral signal pretreatment improved the performance of the PLS. • Variable selection methods can be used to enhance model performance. • CARS-PLS produced optimal results for predicting phenolic contents. [ABSTRACT FROM AUTHOR]

This study aimed to investigate the potential of FT-NIR spectroscopy technique combined with chemometrics method, which employed to monitor time-related changes of alcohol concentration and residual glucose during solid state fermentation (SSF) of ethanol. Characteristic wavelength variables were firstly selected by use of L1-norm regularization approach. Then, the partial least squares (PLS) regression model was finally developed using the variables selected by L1-norm regularization method to quantitative determine alcohol concentration and residual glucose in SSF of ethanol. Compared with the best results of full-spectrum PLS, the L1-PLS model obtained better results as follows: RMSECV = 1.0392 g/L, R c  = 0.9911, RMSEP = 1.0910 g/L, R p  = 0.9917 for alcohol concentration; RMSECV = 1.7002 g/L, R c  = 0.9880, RMSEP = 2.1859 g/L, R p  = 0.9896 for residual glucose. The overall results sufficiently demonstrate that FT-NIR spectroscopy technique coupled with appropriate chemometrics method is a promising tool for monitoring the process of SSF of ethanol. [ABSTRACT FROM AUTHOR]

Volatile organic compounds (VOCs) detection and measurement in materials with near infrared spectroscopy (NIRS) have been an unresolved constraint till date. This paper focused on the use of NIRS for rapid detection and quantification of pure VOCs (ethanol, ethyl acetate and acetic acid) in mixed VOCs via employing sensitive intermediary chemo-responsive dyes as capture probes, whose NIRS spectra were scanned, preprocessed and used to build partial least squares (PLS) prediction models. Average predicted rates based on the PLS-built prediction models for the pure VOCs in the mixed VOCs yielded 98.60 ± 17.41%. 78.26% of the pure VOCs prediction rates ranged between 85 and 114% and normally distributed. The high prediction rates achieved imply the technique may be deployed as a panacea to widen the usage scope of NIRS and e-nose based colorimetric sensors for rapid detection and quantification of VOCs content in materials which hitherto had been a constraint for both systems. [ABSTRACT FROM AUTHOR]

[Display omitted] • Colorimetric sensor with spectral was attempted to classify different grades matcha. • Different classification methods were compared in developing identification models. • Back-propagation artificial neural network model had optimal prediction performance. • Volatile compounds differences between matcha grades were verified by HS-SPME-GC–MS. Matcha tea powder is considered as an integral part of a healthy diet due to its enormous health benefits. In the current study, visible near-infrared (Vis-NIR) and colorimetric sensor array (CSA) techniques are applied to identify the matcha grades. The color-sensitive dyes reacted with their volatile compounds and the information was recorded by Vis-NIR spectroscopy, namely Vis-NIR-CSA. Specifically, three linear and three nonlinear classification models were compared, yielding the optimal identification rate by the back-propagation artificial neural network (BPANN) model with 99% and 98% in the calibration and prediction sets, respectively. The results indicated the sensor combined with the BPANN model could be applied satisfactorily in identification of different matcha grades. Additionally, the variations in volatile compounds between different matcha grades and eight characteristic volatile compounds were identified, which verified the sensor identification results. This study provided a scientific and novel method for the stability of matcha quality in production. [ABSTRACT FROM AUTHOR]

• Chlorophyll during Tencha processing was predicted by portable NIR system. • Wavelength selection algorithms improved the performance of models. • CARS-PLS models showed the best results in modeling. Monitoring chlorophyll during Tencha (the raw ingredient for matcha) processing is critical for determining the matcha's color and quality. The purpose of this study is to explore the mechanism of chlorophyll changes during Tencha processing and evaluate the viability of predicting its content by a portable near-infrared (NIR) spectrometer. The Tencha samples' spectral data were first preprocessed using various preprocessing techniques. Subsequently, iteratively variable subset optimization (IVSO), bootstrapping soft shrinkage (BOSS), and competitive adaptive reweighted sampling (CARS) were used to optimize and build partial least square (PLS) models. The CARS-PLS models achieved the best predictive accuracy, with correlation coefficients of prediction (R p) = 0.9204 for chlorophyll a , R p = 0.9282 for chlorophyll b , and R p = 0.9385 for total chlorophyll. These findings suggest that NIR spectroscopy could be used as a surrogate for immediate quantification and monitoring of chlorophyll during Tencha processing. [ABSTRACT FROM AUTHOR]

[Display omitted] • Total flavonoids and phenolics content were evaluated by NIR. • Preprocessing techniques had no effect on raw data. • Total flavonoids and phenolic prediction were improved with variable selection algorithms. • Si-CARS-PLS exhibited optimal results for flavonoids and phenolic content prediction. Peanuts are nutritionally valuable for both humans and animals due to their high content of flavonoids and phenolic compounds. Herein, we explored the potential of near-infrared (NIR) spectroscopy coupled with efficient variable selection algorithms for quantitative prediction of total flavonoids (TFC) and total phenolics content (TPC) in raw peanut seeds. Spectrophotometrically, the reference results of the extracts for TFC and TPC were analysed and recorded. The integrated application of the synergy interval coupled competitive adaptive reweighted sampling-partial least squares (Si-CARS-PLS) were used for prediction. The model performance appraisal was based on the correlation coefficients of prediction (Rp), root mean square error of prediction (RMSEP), and residual predictive deviation (RPD). The Si-CARS-PLS performed optimally for TFC (Rp = 0.9137, RPD = 2.49) and TPC (Rp = 0.9042, RPD = 2.31), respectively. Moreover, the model (Si-CARS-PLS) was found to have an acceptable fit for the analytes under study since it achieved 0.88 for TFC and 0.86 for TPC based on the external validation. Therefore, these results showed that NIR coupled with Si-CARS-PLS could be used for the quantitative prediction of flavonoids and phenolic contents in raw peanut seeds. [ABSTRACT FROM AUTHOR]