Fourier Transform Near-infrared Research Articles

Application of flash GC e-nose and FT-NIR combined with deep learning algorithm in preventing age fraud and quality evaluation of pericarpium citri reticulatae

Pericarpium citri reticulatae (PCR) is the dried mature fruit peel of Citrus reticulata Blanco and its cultivated varieties in the Brassicaceae family. It can be used as both food and medicine, and has the effect of relieving cough and phlegm, and promoting digestion. The smell and medicinal properties of PCR are aged over the years; only varieties with aging value can be called “Chenpi”. That is to say, the storage year of PCR has a great influence on its quality. As the color and smell of PCR of different storage years are similar, some unscrupulous merchants often use PCRs of low years to pretend to be PCRs of high years, and make huge profits. Therefore, we did this study with the aim of establishing a rapid and nondestructive method to identify the counterfeiting of PCR storage year, so as to protect the legitimate rights and interests of consumers. In this study, a classification model of PCR was established by e-eye, flash GC e-nose, and Fourier transform near-infrared (FT-NIR) combined with machine learning algorithms, which can quickly and accurately distinguish PCRs of different storage years. DFA and PLS-DA models were established by flash GC e-nose to distinguish PCRs of different ages, and 8 odor components were identified, among which (+)-limonene and γ-terpinene were the key components to distinguish PCRs of different ages. In addition, the classification and calibration model of PCRs were established by the combination of FT-NIR and machine learning algorithms. The classification models included SVM, KNN, LSTM, and CNN-LSTM, while the calibration models included PLSR, LSTM, and CNN-LSTM. Among them, the CNN-LSTM model built by internal capsule had significantly better classification and calibration performance than the other models. The accuracy of the classification model was 98.21 %. The R2P of age, (+)-limonene and γ-terpinene was 0.9912, 0.9875 and 0.9891, respectively. These results showed that the combination of flash GC e-nose and FT-NIR combined with deep learning algorithm could quickly and accurately distinguish PCRs of different ages. It also provided an effective and reliable method to monitor the quality of PCR in the market.

Development of a Simultaneous Quantification Method for Multiple Modes of Nitrogen in Leaf Models Using Near-Infrared Spectroscopic Measurement.

By focusing our attention on nitrogen components in plants, which are important for cultivation management in data-driven agriculture, we developed a simple, rapid, non-chemical and simultaneous quantification method for proteinic and nitrate nitrogen in a leaf model based on near-infrared (NIR) spectroscopic information obtained using a compact Fourier Transform NIR (FT-NIR) spectrometer. The NIR spectra of wet leaf models impregnated with a protein-nitric acid mixed solution and a dry leaf model obtained by drying filter paper were acquired. For spectral acquisition, a compact MEMS (Micro Electro Mechanical Systems) FT-NIR spectrometer equipped with a diffuse reflectance probe accessory was used. Partial least square regression analysis was performed using the spectral information of the extracted absorption bands based on the determination coefficients between the spectral absorption intensities and the contents of the two-dimensional spectral analysis between NIR and mid-infrared spectral information. Proteinic nitrogen content in the dry leaf model was well predicted using the MEMS FT-NIR spectroscopic method. Additionally, nitrate nitrogen in the dry leaf model was also determined by the provided method, but the necessity of adding the data for a wider range of nitric acid concentrations was experimentally indicated for the prediction of nitrate nitrogen content in the wet leaf model. Consequently, these results experimentally suggest the possibility of the application of the compact MEMS FT-NIR for obtaining the bioinformation of crops at agricultural on-sites.

Rapid testing in the food industry: the potential of Fourier transform near-infrared (FT-NIR) spectroscopy and spatially offset Raman spectroscopy (SORS) to detect raw material defects in hazelnuts (Corylus avellana L.)

The detection of raw material defects, e.g., due to incorrect or excessively long storage, is an important issue in incoming goods inspections in the food industry. Fast and easy-to-use analytical methods for evaluating the usability of raw materials are particularly important. In this study, the applicability of Fourier transform near-infrared (FT-NIR) spectroscopy and spatially offset Raman spectroscopy (SORS) for the detection of raw material defects was evaluated. For this purpose, six hazelnut batches stored at different temperatures, humidity levels, and storage times were used as examples in this pilot study. Classification models of samples before and after the different physical treatments show that the resulting changes can be detected by FT-NIR spectroscopy and SORS at elevated temperature and humidity. When one of the storage parameters is increased, FT-NIR spectroscopy is also useful for detecting differences between sample groups. In contrast, SORS cannot distinguish between pre- and post-stored samples when only one of the storage parameters is increased, making SORS unsuitable for incoming inspection of nuts. FT-NIR spectroscopy analysis is also a fast application, because freeze-drying of the sample material prior to analysis is not required as the results before and after freeze-drying are comparable. Combining the FT-NIR spectroscopy and SORS data in a low-level data fusion improved the classification models for samples stored at low storage temperatures, suggesting that the two methods provide complementary information. In summary, analyzing nuts with FT-NIR spectroscopy and SORS, as shown for hazelnuts, has the potential to identify abnormal samples during incoming goods inspections.

Quantitative Analysis of Near-Infrared Spectroscopy Using the BEST-1DConvNet Model

In the quest for enhanced precision in near-infrared spectroscopy (NIRS), in this study, the application of a novel BEST-1DConvNet model for quantitative analysis is investigated against conventional support vector machine (SVM) approaches with preprocessing such as multiplicative scatter correction (MSC) and standard normal variate (SNV). We assessed the performance of these methods on NIRS datasets of diesel, gasoline, and milk using a Fourier Transform Near-Infrared (FT-NIR) spectrometer having a wavelength range of 900–1700 nm for diesel and gasoline and 4000–10,000 nm for milk, ensuring comprehensive spectral capture. The BEST-1DConvNet’s effectiveness in chemometric predictions was quantitatively gauged by improvements in the coefficient of determination (R2) and reductions in the root mean square error (RMSE). The BEST-1DConvNet model achieved significant performance enhancements compared to the MSC + SNV + 1D + SVM model. Notably, the R2 value for diesel increased by approximately 48.85% despite a marginal RMSE decrease of 0.92%. R2 increased by 11.30% with a 3.32% RMSE reduction for gasoline, and it increased by 8.71%, accompanied by a 3.51% RMSE decrease for milk. In conclusion, the BEST-1DConvNet model demonstrates superior predictive accuracy and reliability in NIRS data analysis, marking a substantial leap forward in spectral analysis technology. This advancement could potentially streamline their integration into various industrial applications and highlight the role of convolutional neural networks in future chemometric methodologies.

Analysis of Polyphenol Patterns of Pleurotus ostreatus Cultivars by UHPLC-ESI-MS/MS; Application of FT-NIR and Chemometric Methods, Classification Options

The nutritional properties of Pleurotus mushrooms were studied to select the varieties with the most favourable properties. These mushrooms have high nutritional value; they are rich in carbohydrates, protein, minerals, vitamins, chitin and reducing compounds, such as phenols and polyphenols. In this study, the polyphenol profiles of thirteen Pleurotus ostreatus cultivars were established by the UHPLC-ESI-MS/MS technique. The results showed that 4-hydroxibenzoic acid, caffeic acid, p-coumaric acid and vanillic acid were the most abundant polyphenolic components in the samples. In addition, the Fourier-transformed near infrared (FT-NIR) spectra of the samples were recorded and evaluated. The correlation between the differences in NIR spectra and the differences in polyphenol patterns of the samples was investigated. The polyphenol results were subjected to several statistical evaluations (Kruskal–Wallis test, Principal Component Analysis (PCA), Spearman correlation analysis, cluster analysis) to detect possible differences between the samples. Relationships between the polyphenol profile and antioxidant capacity (FRAP), total polyphenol content (TPC), free amino acid content (fAA) and the values of each polyphenol component were examined. Based on the results, an effort was made to group the varieties according to the attributes tested.

Rapid and Non-Destructive Determination of Fatty Acid Profile and Oil Content in Diverse Brassica carinata Germplasm Using Fourier-Transform Near-Infrared Spectroscopy

Brassica carinata is one of the oilseeds in the Brassicaceae family, possessing seed quality traits such as oil with various fatty acid profiles suitable for many industrial applications. Determination of such quality traits using conventional methods is often expensive, time-consuming, and destructive. In contrast, the Near-Infrared Spectroscopic (NIRS) technique has been proven fast, cost-effective, and non-destructive for the determination of seed compositions. This study aimed to demonstrate that NIRS is a rapid and non-destructive method for determining the fatty acid profile and oil content in diverse germplasms of B. carinata. A total of 96 genetically diverse B. carinata germplasms that include accessions, advanced breeding lines, and varieties were used in this study. Reference data sets were generated using gas chromatography and the Soxhlet oil extraction method for fatty acid profile and oil content, respectively. Spectra data were taken from the wavenumber range of 11,500 to 4000 cm−1 using the Fourier-transform near-infrared (FT-NIR) method. NIRS calibration equations were developed using partial least square (PLS) regression with OPUS software, version 7.5.1. Higher coefficient of determination (R2val) and ratio of performance to deviation (RPD) > 3 were obtained for oleic acid (R2val = 0.92, RPD = 3.6), linoleic acid (R2val = 0.89, RPD = 3.2), linolenic acid (R2val = 0.93, RPD = 3.8), erucic acid (R2val = 0.92, RPD = 3.5), and oil content (R2val = 0.93, RPD = 3.6). Thus, the NIRS calibration models for the aforementioned fatty acids and oil content were found to be strong enough for prediction. However, the calibration models for palmitic acid (R2val = 0.78, RPD = 2.1) and stearic acid (R2val = 0.75, RPD = 2.0) showed relatively smaller R2val and thus became weaker in their prediction capacity. Despite their relatively lower R2, the calibration equations for palmitic and stearic acids could be used for approximate estimation and rough screening purposes. In conclusion, the calibration models that we have developed will be useful in applying NIRS as a high-throughput, non-destructive method for the screening of large germplasms in terms of their fatty acid profiles and oil content during the oil quality breeding efforts conducted on B. carinata.

Otolith morphometry and Fourier transform near-infrared (FT-NIR) spectroscopy as tools to discriminate archived otoliths of newly detected cryptic species, Etelis carbunculus and Etelis boweni

Cryptic speciation was recently verified in Etelis carbunculus, an important component of federally managed bottomfish fisheries in the Pacific Territories of the United States. As a result, archived otolith collections used for fishery assessment are now contaminated with newly described E. boweni in areas where these species co-occur. We compared the efficacy of otolith morphometrics and Fourier transform near-infrared (FT-NIR) spectroscopy to discriminate species first using voucher (i.e., known species) otoliths (n = 93) from the SW Pacific, then applied optimal models to archived otoliths (n = 91) collected around Guam. Significant and distinguishable differences in otolith morphometrics as well as FT-NIR spectral absorbance patterns were observed between E. carbunculus and E. boweni voucher samples. Classification models applied using both morphometric measurements (quadratic discriminant analysis) and FT-NIR spectral data (partial least squares discriminant analysis) were able to predict species with a high (93 – 100%) degree of accuracy despite a relatively large spatial area of specimen collection ( ± 10° latitude and longitude) and regardless of whether otoliths were whole (i.e., unbroken). Further, each method identified members of newly described E. boweni in the archived collection of E. carbunculus otoliths captured around Guam, providing strong evidence that the species’ distributions overlap in this region. The purported identification of both E. carbunculus and E. boweni in the archived catch from Guam has important implications for fisheries management; therefore, it is imperative that the corresponding otolith collections are examined to ensure that the otoliths are assigned to the correct species.

Effect of drying temperature on composition of edible mushrooms: Characterization and assessment via HS-GC-MS and IR spectral based volatile profiling and chemometrics

Edible wild mushrooms are one of the popular ingredients due to their high quality and unique flavor and nutrients. To gain insight into the effect of drying temperature on its composition, 86 Boletus bainiugan were divided into 5 groups and dried at different temperatures. Headspace solid-phase microextraction-gas chromatography-mass spectrometry (HS-SPME-GC-MS) was used for the identification of volatile organic compounds (VOCs) of Boletus bainiugan. The 21 differential VOCs that distinguish different drying temperatures of Boletus bainiugan were identified. 65 °C retained more VOCs. There were differences in their types and content at different temperatures, proteins, polysaccharides, crude fibers, and fats. Fourier transform near-infrared (FT-NIR) spectroscopy, Fourier transform infrared (FTIR) spectroscopy, and two-dimensional correlation spectroscopy (2DCOS) images were successfully characterized for differences in the chemical composition of Boletus bainiugan. Partial least squares discriminant analysis (PLS-DA) verified the variability in the chemical composition of Boletus bainiugan with the coefficient of determination (R2) = 0.95 and predictive performance (Q2) = 0.75 with 92.31% accuracy. Next, infrared spectroscopy provides a fast and efficient assessment of the content of Boletus bainiugan nutrients (proteins, polysaccharides, crude fibers, and fats).

Rapid detection of soil carbonates by means of NIR spectroscopy, deep learning methods and phase quantification by powder X-ray diffraction

In this study we propose a novel rapid and efficient way to predict carbonates content in soil by means of Fourier Transform Near-Infrared (FT-NIR) reflectance spectroscopy and by use of deep learning methods. In addition to using traditional machine learning algorithms, we exploited multiple deep learning methods, such as: 1) a Multi-Layered Perceptron Regressor (MLP) and 2) a Convolutional Neural Network (CNN) in an attempt to compare their performance with other classical machine learning algorithms, which up until now were considered the field’s standards, such as Partial Least Squares Regression (PLSR), Cubist and SupportVectorMachines (SVM) on the combined dataset of two NIR spectral libraries: Kellogg Soil Survey Laboratory (KSSL) of the United States Department of Agriculture (USDA), a dataset of soil samples reflectance spectra collected nationwide, and Land Use and Coverage Area Frame Survey (LUCAS) TopSoil-2015 (European Soil Library) which contains soil sample absorbance spectra from all over the European Union, and use them to predict carbonate content on never-before-seen soil samples. In this study, absorbances in the NIR spectral region (1150-2500 nm) were utilized in two different ways: a) as one-dimensional data and b) as two-dimensional spectrograms which, to our knowledge, is a completely novel approach that has rarely been researched. Quantification of carbonates by means of X-ray-Diffraction is in good agreement with the volumetric method and the MLP prediction. Our work contributes to rapid carbonates content prediction in soil samples in cases where: 1) no volumetric method is available and 2) only NIR spectra absorbance data are available.

A Novel Variable Selection Method Based on Ordered Predictors Selection and Successive Projections Algorithm for Predicting Gastrodin Content in Fresh Gastrodia elata Using Fourier Transform Near-Infrared Spectroscopy and Chemometrics.

Gastrodin is one of the most important biologically active components of Gastrodia elata, which has many health benefits as a dietary and health food supplement. However, gastrodin measurement traditionally relies on laboratory and sophisticated instruments. This research was aimed at developing a rapid and non-destructive method based on Fourier transform near infrared (FT-NIR) to predict gastrodin content in fresh Gastrodia elata. Auto-ordered predictors selection (autoOPS) and successive projections algorithm (SPA) were applied to select the most informative variables related to gastrodin content. Based on that, partial least squares regression (PLSR) and multiple linear regression (MLR) models were compared. The autoOPS-SPA-MLR model showed the best prediction performances, with the determination coefficient of prediction (Rp2), ratio performance deviation (RPD) and range error ratio (RER) values of 0.9712, 5.83 and 27.65, respectively. Consequently, these results indicated that FT-NIRS technique combined with chemometrics could be an efficient tool to rapidly quantify gastrodin in Gastrodia elata and thus facilitate quality control of Gastrodia elata.

E-eye and FT-NIR combined with multivariate algorithms to rapidly evaluate the dynamic changes in the quality of Gastrodia elata during steaming process

Gastrodia elata (GE) is traditionally subjected to steaming, and steaming duration plays a crucially important role in determining GE quality. This study examined the variations in bioactive components during the steaming process and proposed the utilization of electronic eye and Fourier Transform near-infrared (FT-NIR) spectroscopy for quality assessment. The findings revealed that the levels of parishin E parishin B, parishin A, and gastrodin initially rose and subsequently declined, while 4-Hydroxybenzyl alcohol exhibited a rapid decrease followed by stabilization. With prolonged steaming, the brightness of GE decreased, while the red and yellow tones became more pronounced and the color saturation increased. FT-NIR divided the steaming process into three stages: 0 min (raw GE), 0–9 min (partially steamed GE), and 9–30 min (fully steamed GE), and the partial least squares regression models effectively predicted the levels of five components. Overall, this study provided valuable insights into quality control in food processing.

Rapid evaluation of Ziziphi Spinosae Semen and its adulterants based on the combination of FT-NIR and multivariate algorithms

Ziziphi Spinosae Semen (ZSS) is a valued seed renowned for its sedative and sleep-enhancing properties. However, the price increase has been accompanied by adulteration. In this study, chromaticity analysis and Fourier transform near-infrared (FT-NIR) combined with multivariate algorithms were employed to identify the adulteration and quantitatively predict the adulteration ratio. The findings suggested that the utilization of chromaticity extractor was insufficient for identification of adulteration ratio. The raw spectrum of ZMS and HAS adulterants extracted by FT-NIR was processed by SNV + CARS and 1d + SG + ICO respectively, the average accuracy of machine learning classification model was improved from 77.06 % to 97.58 %. Furthermore, the R2 values of the calibration and prediction set of the two quantitative prediction regression models of adulteration ratio are greater than 0.99, demonstrating excellent linearity and predictive accuracy. Overall, this study demonstrated that FT-NIR combined with multivariate algorithms provided a significant approach to addressing the growing issue of ZSS adulteration.

A Comparative Study of Fourier Transform Near-Infrared Spectroscopy and Physicochemical Methods in Wheat Quality Analysis

Near-infrared spectroscopy is a non-invasive, rapid, and efficient analytical method widely employed in agricultural quality assessment. This study aims to explore the potential of Fourier transform near-infrared (FT-NIR) spectroscopy in the analysis of wheat quality in comparison to traditional physicochemical methods. Key quality parameters such as protein content, bulk density, dough extensibility, development and stability time, wet gluten, and extension area were considered. To compare the similarities between the two methods in an irregular manner and to make the trials more representative, randomized combinations of different samples were made by planting different wheat genotypes in different locations. In wheat protein determination (r = 0.92, p < 0.001), the correlation between protein determinations was very high and highly statistically significant. This indicates that the Fourier transform near-infrared (FT-NIR) assay is highly consistent and accurate with the physicochemical method for protein determination. In the quality indices of bulk density (r = 0.46, p < 0.05), dough extensibility (r = 0.46, p < 0.05), development time (r = 0.37, p < 0.05), stability time (r = 0.49, p < 0.01), wet gluten (r = 0.54, p < 0.01), and dough extension area (r = 0.63, p < 0.001), there was a positive correlation between the two assays. Although the correlation was lower compared to the protein assay, it was still statistically significant. This suggests that the FT-NIR assay also has some accuracy in these indices, although there may be some differences from the physicochemical method. Through comparative analysis, we found that in wheat quality assessment, FT-NIR showed a strong correlation and was highly significant in the determination of wheat protein content; in the determination of the extension area, although highly significant, the correlation coefficient was not high, and there was a positive correlation between the two, and a lower correlation was shown in the bulk density, dough extensibility, and development time. These results indicate that FT-NIR can assess the wheat protein content quality indicator. However, its ability to accurately assess wheat quality indicators such as density, dough extensibility, development and stability time, wet gluten, and dough extensibility needs further investigation.

Primary assessment of macronutrients in durian (CV Monthong) leaves using near infrared spectroscopy with wavelength selection

Farmers would be able to regulate fertilization and produce quality durian if they knew the nutrient concentration in durian leaves. A long period of time for traditional nutritional content determination is needed. Therefore, near-infrared spectroscopy is a good method for nondestructive and quick nutrient content evaluation. The leaf sample matrices (fresh leaves, dried ground leaves, and dried ground leaf pellets) were scanned by Fourier transform near-infrared (FT-NIR) with a wavelength of 12,500–3,600 cm−1. Regression models were developed using partial least squares (PLS) with full wavelength, short wavelength, and selected wavelength by successive projections algorithm (SPA). In this study, the model for N and K concentration was acceptable and the prediction was considered good but for P content not had succeeded. As a result, the PLS-SPA model using fresh leaf samples for evaluating N content in durian leaves exhibited performance of r2 = 0.852, SEP = 0.14%, RPD = 2.63 and bias = −0.020%. The PLS-SPA model using dried ground leaf samples for evaluating K content in durian leaves exhibited performance of r2 = 0.820, SEP = 0.13%, RPD = 2.36 and bias = 0.006%. This research found that it is possible to apply NIR waves to predict N and K concentrations in durian leaves. It is not necessary to predict directly from the wavelengths associated with -N or -K bonds. Instead, NIR can measure them indirectly from the bonding of proteins, which are products formed by N and K. In addition, selecting the wavelength that is related to the value to be measured can produce results that are not significantly different from using full or short wavelengths. These models can assist farmers in rapidly predicting N and K content in durian leaves for immediate fertilizer adjustment.

Oregano herb adulteration detection through rapid spectroscopic approaches: Fourier transform-near infrared and laser photoacoustic spectroscopy facilities

Oregano is a culinary herb known and used worldwide, whose demand and cost encourage illegal activities in terms of fraud, especially adulteration. The scope of this study is to provide a rapid, cost-effective and easy-to-use approach to detect the adulteration of oregano with a common matrix, the olive leaves. Two analytical solutions were considered, Fourier transform near infrared (FT-NIR), using two different instruments, and laser photoacoustic spectroscopy (LPAS). Different raw data preprocessing procedures were applied, Principal Component Analysis (PCA) was used for preliminary data visualisation, while Partial Least Squares Regression (PLS-R) and Discriminant Analysis (PLS-DA) allowed the authors to classify samples with different adulteration percentages. FT-NIR performed better than LPAS, perhaps also because the photoacoustic system is still an experimental prototype. For example, the coefficient of determination R2 between actual and predicted adulterant concentrations was in the range 0.93–1.00 for FT-NIR and in the range 0.86–0.96 for LPAS. Nevertheless, this allowed the use of a low level data fusion approach to increase the consistency of the statistical model. The fusion between raw data from the LPAS prototype and one of the FT-NIR instruments gave optimal results, confirming the potential of these analytical solutions for the identification of oregano adulterants. The approaches used in the present work could be applied to the identification of other adulterants in herbs and spices by food companies for quality control measurements, possibly for on line and in field analysis.

Rapid Determination of Nutmeg Shell Content in Ground Nutmeg Using FT-NIR Spectroscopy and Machine Learning.

Nutmeg is a popular spice often used in ground form, which makes it highly susceptible to food fraud. Therefore, the aim of the present study was to detect adulteration of ground nutmeg with nutmeg shell via Fourier transform near-infrared (FT-NIR) spectroscopy. For this purpose, 36 authentic nutmeg samples and 10 nutmeg shell samples were analyzed pure and in mixtures with up to 50% shell content. The spectra plot as well as a principal component analysis showed a clear separation trend as a function of shell content. A support vector machine regression used for shell content prediction achieved an R2 of 0.944 in the range of 0-10%. The limit of detection of the prediction model was estimated to be 1.5% nutmeg shell. Based on random sub-sampling, the likelihood was found to be 2% that a pure nutmeg sample is predicted with a nutmeg shell content of >1%. The results confirm the suitability of FT-NIR spectroscopy for rapid detection and quantitation of the shell content in ground nutmeg.

Detection of the Early Fungal Infection of Citrus by Fourier Transform Near-Infrared Spectra

Early fungal infection of citrus is one of the common diseases found during the storage period of citrus, and fungus that infects citrus will spread to the entire batch of citrus as the degree of infection deepens, causing enormous economic losses. Therefore, early detection of fungal infection of citrus is fundamental. The purpose of this study is to explore the qualitative identification of early fungal infections in citrus by using Fourier transform near-infrared (FT-NIR) combined with a variety of chemometric methods. First, discrete wavelet transform (DWT) is used to filter the noise of the spectral signal, then combined with a PLS-DA model, that helps discriminate healthy from infected Citrus. Subsequently, four different feature variable selection methods were introduced, Then, the linear discriminant analysis (LDA) and support vector machine (SVM) two classifiers were combined to establish a qualitative model for the degree of fungal infection. The modeling results show that the SVM modeling effect is better than LDA, and the DWT-CARS-SVM based on the RBF kernel function has the best result, the accuracy rates of the training set and test set are 100% and 97%. The results indicate that FT-NIR spectroscopy, combined with chemometric methods, is able to distinguish early fungal infections in citrus.

Detection of Camellia oleifera anthracnose based on THz combined with FT-NIR

Camellia oleifera anthracnose is a prevalent and highly destructive disease in Camellia oleifera industry, seriously restricting its development. In light of the current problems of complex and inefficient detection of Camellia oleifera anthracnose, data fusion technology has been widely employed across various fields. Extensive research has demonstrated that data fusion enhances the detection accuracy and classification accuracy of the model. Consequently, a data fusion method combining terahertz spectroscopy (THz) and Fourier transform near-infrared (FT-NIR) spectroscopy was proposed to detect the level of Camellia oleifera anthracnose. Partial least squares discriminant analysis (PLS-DA) was utilized to establish the single-spectrum and fusion-spectrum anthracnose classification models of Camellia oleifera, enabling rapid, efficient, non-destructive, and highly precise determination of Camellia oleifera anthracnose. The single spectral model, THz-PLS-DA, exhibited a misjudgment rate of 7.5 % in the modeling set, and 17.5 % in the prediction set. In contrast, the low-level fusion model, THz-FT-NIR-PLS-DA, exhibited a misjudgment rate of 3.25 % in the modeling set and 0 in the prediction set. The intermediate fusion model, (THz-VIP)-(FT-NIR-VIP)-PLS-DA, showcased a modeling set misjudgment rate of 0.75 % and a prediction set misjudgment rate of 0. Importantly, the intermediate-level model exhibited superior qualitative analysis accuracy and stability compared to both the single model and the low-level fusion model. THz spectroscopy combined with FT-NIR spectroscopy can be used to nondestructively, rapidly and accurately differentiate healthy Camellia oleifera from different grades of anthracnose Camellia leaves.

Synchronously Predicting Tea Polyphenol and Epigallocatechin Gallate in Tea Leaves Using Fourier Transform-Near-Infrared Spectroscopy and Machine Learning.

Tea polyphenol and epigallocatechin gallate (EGCG) were considered as key components of tea. The rapid prediction of these two components can be beneficial for tea quality control and product development for tea producers, breeders and consumers. This study aimed to develop reliable models for tea polyphenols and EGCG content prediction during the breeding process using Fourier Transform-near infrared (FT-NIR) spectroscopy combined with machine learning algorithms. Various spectral preprocessing methods including Savitzky-Golay smoothing (SG), standard normal variate (SNV), vector normalization (VN), multiplicative scatter correction (MSC) and first derivative (FD) were applied to improve the quality of the collected spectra. Partial least squares regression (PLSR) and least squares support vector regression (LS-SVR) were introduced to establish models for tea polyphenol and EGCG content prediction based on different preprocessed spectral data. Variable selection algorithms, including competitive adaptive reweighted sampling (CARS) and random forest (RF), were further utilized to identify key spectral bands to improve the efficiency of the models. The results demonstrate that the optimal model for tea polyphenols calibration was the LS-SVR with Rp = 0.975 and RPD = 4.540 based on SG-smoothed full spectra. For EGCG detection, the best model was the LS-SVR with Rp = 0.936 and RPD = 2.841 using full original spectra as model inputs. The application of variable selection algorithms further improved the predictive performance of the models. The LS-SVR model for tea polyphenols prediction with Rp = 0.978 and RPD = 4.833 used 30 CARS-selected variables, while the LS-SVR model build on 27 RF-selected variables achieved the best predictive ability with Rp = 0.944 and RPD = 3.049, respectively, for EGCG prediction. The results demonstrate a potential of FT-NIR spectroscopy combined with machine learning for the rapid screening of genotypes with high tea polyphenol and EGCG content in tea leaves.

Quantitative analysis of residues of chlorpyrifos in corn oil based on Fourier transform near-infrared spectroscopy and deep transfer learning

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 (RP2) 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.