Visible/Near-Infrared Spectra for Linear and Nonlinear Calibrations: A Case to Predict Soluble Solids Contents and pH Value in Peach

Two sensitive wavelength (SWs) selection methods combined with visible/near-infrared (Vis/NIR) spectroscopy were investigated to determine the soluble solids content (SSC) and pH value in peaches, including latent variables analysis (LVA) and independent component analysis (ICA). A total of 100 samples were prepared for the calibration (n = 70) and prediction (n = 30) sets. Calibration models using SWs selected by LVA and ICA were developed, including linear regression of partial least squares (PLS) analysis and nonlinear regression of least squares-support vector machine (LS-SVM). In the nonlinear models, four SWs selected by ICA achieved the optimal ICA-LS-SVM model compared with LV-LS-SVM and both of them better than linear model of PLS. The correlation coefficients (r _p and r _cv), root mean square error of cross validation, root mean square error of prediction, and bias by ICA-LS-SVM were 0.9537, 0.9485, 0.4231, 0.4155, and 0.0167 for SSC and 0.9638, 0.9657, 0.0472, 0.0497, and −0.0082 for pH value, respectively. The overall results indicated that ICA was a powerful way for the selection of SWs, and Vis/NIR spectroscopy incorporated to ICA-LS-SVM was successful for the accurate determination of SSC and pH value in peach.     

Infrared spectroscopy and chemometrics for the starch and protein prediction in irradiated rice

Infrared spectroscopy was investigated to predict components of starch and protein in rice treated with different irradiation doses based on sensitive wavelengths (SWs). Near infrared and mid-infrared regions were compared to determine which one produces the best prediction of components in rice after irradiation. Partial least-squares (PLS) analysis and least-squares-support vector machine (LS-SVM) were implemented for calibration models. The best PLS models were achieved with NIR region for starch and MIR region for protein. Moreover, independent component analysis (ICA) was executed to select several SWs based on loading weights, and the optimal LS-SVM model was achieved with SWs of 1210–1222, 1315–1330, 1575–1625, 1889–1909 and 2333–2356 nm for starch and SWs of 962–1091, 1232–1298, 1480–1497, 1584–1625 and 2373–2398 cm⁻¹ for protein. It indicated that IR spectroscopy combined with LS-SVM could be applied as a high precision way for the determination of starch and protein in rice after irradiation.     

A comparative study for the quantitative determination of soluble solids content,pH and firmness of pears by Vis/NIR spectroscopy

Visible and near infrared (Vis/NIR) spectroscopy was investigated to determine the soluble solids content (SSC), pH and firmness of different varieties of pears. Two-hundred forty samples (80 for each variety) were selected as sample set. Two-hundred ten pear samples (70 for each variety) were selected randomly for the calibration set, and the remaining 30 samples (10 for each variety) for the validation set. Partial least squares (PLS) and least squares-support vector machine (LS-SVM) with different spectral preprocessing techniques were implemented for calibration models. Different wavelength regions including Vis, NIR and Vis/NIR were compared. It indicated that Vis/NIR (400–1800 nm) was optimal for PLS and LS-SVM models. Then, LS-SVM models were developed with a grid search technique and RBF kernel function. All LS-SVM models outperformed PLS models. Next, effective wavelengths (EWs) were selected according to regression coefficients. The EW-LS-SVM models were developed and a good prediction precision and stability was achieved compared with PLS and LV-LS-SVM models. The correlation coefficient of prediction (r_p), root mean square error of prediction (RMSEP) and bias for the best prediction by EW-LS-SVM were 0.9164, 0.2506 and −0.0476 for SSC, 0.8809, 0.0579 and −0.0025 for pH, whereas 0.8912, 0.6247 and −0.2713 for firmness, respectively. The overall results indicated that the regression coefficient was an effective way for the selection of effective wavelengths. LS-SVM was superior to the conventional linear PLS method in predicting SSC, pH and firmness in pears. Therefore, non-linear models may be a better alternative to monitor internal quality of fruits. And the EW-LS-SVM could be very helpful for development of portable instrument or real-time monitoring of the quality of pears.     

Applying Near-Infrared Spectroscopy and Chemometrics to Determine Total Amino Acids in Herbicide-Stressed Oilseed Rape Leaves

Near-infrared (NIR) spectroscopy was investigated to determine the total amino acids (TAA) in oilseed rape (Brassica napus L.) leaves under a new herbicide—propyl 4-(2-(4,6-dimethoxypyrimidin-2-yloxy)benzylamino)benzoate (ZJ0273)—stress. In full-spectrum partial least squares (PLS) models, direct orthogonal signal correction (DOSC) was the best preprocessing method. Successive projections algorithm (SPA) was used to select the relevant variables. Multiple linear regression (MLR), PLS, and least squares-support vector machine (LS-SVM) were used for calibration. The DOSC–SPA–LS-SVM model achieved the best prediction performance with correlation coefficients r = 0.9968 and root mean squares error of prediction (RMSEP) = 0.2950 comparing all SPA–MLR, SPA–PLS, and SPA–LS-SVM models. Some parsimonious direct functions were also developed based on the DOSC–SPA wavelength (1,340 nm) such as linear, index, logarithmic, binominal, and exponential functions. The best performance was achieved by direct exponential function with r = 0.9968 and RMSEP = 0.2943. The overall results indicated that NIR was able to determine the TAA in herbicide-stressed oilseed rape leaves, and the DOSC–SPA was quite helpful for the development of detection sensors and the monitoring of the growing status and herbicide effect on field crop oilseed rape.     

Measurement of Soluble Solids Content and pH of Yogurt Using Visible/Near Infrared Spectroscopy and Chemometrics

Visible/near infrared spectroscopy (Vis/NIRs) technique was applied to non-destructive quantification of sugar and pH value in yogurt. Partial least squares (PLS) analysis and least squares support vector machine (LS-SVM) were implemented for calibration models. In this paper, three brands (Mengniu, Junyao, and Guangming) were set as the calibration, and the remaining two brands (Yili and Shuangfeng) were used as prediction set. In the LS-SVM model, the correlation coefficient (r), root mean square error of prediction, and bias in prediction set were 0.9427, 0.2621°Brix, 1.804e−09 for soluble solids content, and 0.9208, 0.0327, and 1.094e−09 for pH, respectively. The correlation spectra corresponding to the soluble solids content and pH value of yogurt were also analyzed through PLS method. LS-SVM model was better than PLS models for the measurements of soluble solids content and pH value. The results showed that the Vis/NIRs combined with LS-SVM models could predict the soluble solids content and pH value of yogurt.     

Rapid detecting total acid content and classifying different types of vinegar based on near infrared spectroscopy and least-squares support vector machine

More than 3.2 million litres of vinegar is consumed every day in China. There are many types of vinegar in China. How to control the quality of vinegar is problem. Near infrared spectroscopy (NIR) transmission technique was applied to achieve this purpose. Ninety-five vinegar samples from 14 origins covering 11 provinces in China were collected. They were classified into mature vinegar, aromatic vinegar, rice vinegar, fruit vinegar, and white vinegar. Fruit vinegar and white vinegar were separated from the other traditional categories in the two-dimension principal component space of NIR after principle component analysis (PCA). Least-squares support vector machine (LS-SVM) as the pattern recognition was firstly applied to identify mature vinegar, aromatic vinegar, rice vinegar in this study. The top two principal components (PCs) were extracted as the input of LS-SVM classifiers by principal component analysis (PCA). The best experimental results were obtained using the radial basis function (RBF) LS-SVM classifier with σ = 0.8. The accuracies of identification were more than 85% for three traditional vinegar categories. Compared with the back propagation artificial neural network (BP-ANN) approach, LS-SVM algorithm showed its excellent generalisation for identification results. As total acid content (TAC) is highly connecting with the quality of vinegar, NIR was used to prediction the TAC of samples. LS-SVM was applied to building the TAC prediction model based on spectral transmission rate. Compared with partial least-square (PLS) model, LS-SVM model gave better precision and accuracy in predicting TAC. The determination coefficient for prediction (R_p) of the LS-SVM model was 0.919 and root mean square error for prediction (RMSEP) was 0.3226. This work demonstrated that near infrared spectroscopy technique coupled with LS-SVM could be used as a quality control method for vinegar.     

Detection of Organic Acids and pH of Fruit Vinegars Using Near-Infrared Spectroscopy and Multivariate Calibration

Near-infrared (NIR) spectroscopy was investigated to determine the acetic, tartaric, formic acids and pH of fruit vinegars. Optimal partial least squares (PLS) models were developed with different preprocessing. Simultaneously, the performance of least squares-support vector machine (LS-SVM) models was compared with three kinds of inputs, including wavelet transform (WT), latent variables, and effective wavelengths (EWs). The results indicated that all LS-SVM models outperformed PLS models. The optimal correlation coefficient (r), root mean square error of prediction and bias for validation set were 0.9997, 0.3534, and −0.0110 for acetic acid by WT-LS-SVM; 0.9985, 0.1906, and 0.0025 for tartaric acid by WT-LS-SVM; 0.9987, 0.1734, and 0.0012 for formic acid by EW-LS-SVM; and 0.9996, 0.0842, and 0.0012 for pH by WT-LS-SVM, respectively. The results indicated that NIR spectroscopy (7,800–4,000 cm⁻¹) combined with LS-SVM could be utilized as a precision method for the determination of organic acids and pH of fruit vinegars.     

Non-destructive measurement of acidity of Chinese bayberry using Vis/NIRS techniques

Chinese bayberry (Myrica rubra Siebold and Zuccarini) is cultivated in southeast China for its edible fruits. In this research, the potential of using the visible/near infrared spectroscopy (Vis/NIRS) was investigated for measuring the acidity of Chinese bayberry, and the relationship was established between non-destructive Vis/NIRS measurement and the acidity of Chinese bayberry. Intact Chinese bayberry fruit was measured by reflectance Vis/NIR in 325–1075 nm range. The data set as the logarithms of the reflectance reciprocal (absorbance (log 1/R)) was analyzed in order to build the best prediction model for this characteristic, using several spectral pretreatments and multivariate calibration techniques such as partial least square regression (PLS). The model for prediction the acidity (r=0.963), standard error of prediction (SEP) 0.21 with a bias of 0.138; showed an excellent prediction performance. The Vis/NIRS technique has significantly greater accuracy for determining the acidity. This non-destructive, fast and accuracy technology can be used in food industry that would be beneficial to human health.     

Discrimination of Producing Areas of Auricularia auricula Using Visible/Near Infrared Spectroscopy

Visible and near infrared (Vis/NIR) spectroscopy combined with chemometric methods was applied for the discrimination of producing areas of Auricularia auricula. Four major varieties of commercial A. auricula were prepared for spectral acquisition. Some pretreatments were performed, such as Savitzky–Golay smoothing, standard normal variate, and the first and second Savitzky–Golay derivative. The scores of the top four latent variables, extracted by partial least squares, were considered as the inputs of back propagation neural network (BPNN) and least squares-support vector machine (LS-SVM). The performance was validated by 60 validation samples. The excellent recognition ratio was 98.3% by BPNN and 96.7% by LS-SVM model with the threshold prediction error ±0.1. The results indicated that Vis/NIR spectroscopy could be used as a rapid and high-precision method for the discrimination of different producing areas of A. auricula by both BPNN and LS-SVM methods.     

Measurement of Soluble Solid Contents and pH of White Vinegars Using VIS/NIR Spectroscopy and Least Squares Support Vector Machine

Visible and near-infrared (VIS/NIR) spectroscopy combined with least squares support vector machine (LS-SVM) was employed to determine soluble solid contents (SSC) and pH of white vinegars. Three hundred twenty vinegar samples were distributed into a calibration set (240 samples) and a validation set (80 samples). Partial least squares (PLS) analysis was implemented for the regression model and extraction of latent variables (LVs). The selected LVs were used as LS-SVM input variables. Finally, LS-SVM models with radial basis function kernel were achieved with the comparison of PLS models. The results indicated that LS-SVM outperformed PLS models. The correlation coefficient (r), root mean square error of prediction, bias, and residual prediction deviation for the validation set were 0.988, 0.207°Brix, 0.183, and 6.4 for SSC whereas these were 0.988, 0.041, ?0.002, and 6.5 for pH, respectively. The overall results indicated that VIS/NIR spectroscopy and LS-SVM could be used as a rapid alternative method for the prediction of SSC and pH of white vinegars, and the results could be helpful for the fermentation process and quality control monitoring of white vinegar production.     

Visible and near-infrared spectroscopic analysis of raw milk for cow health monitoring: reflectance or transmittance?

The composition of produced milk has great value for the dairy farmer. It determines the economic value of the milk and provides valuable information about the metabolism of the corresponding cow. Therefore, online measurement of milk components during milking 2 or more times per day would provide knowledge about the current health and nutritional status of each cow individually. This information provides a solid basis for optimizing cow management. The potential of visible and near-infrared (Vis/NIR) spectroscopy for predicting the fat, crude protein, lactose, and urea content of raw milk online during milking was, therefore, investigated in this study. Two measurement modes (reflectance and transmittance) and different wavelength ranges for Vis/NIR spectroscopy were evaluated and their ability to measure the milk composition online was compared. The Vis/NIR reflectance measurements allowed for very accurate monitoring of the fat and crude protein content in raw milk (R² > 0.95), but resulted in poor lactose predictions (R² < 0.75). In contrast, Vis/NIR transmittance spectra of the milk samples gave accurate fat and crude protein predictions (R² > 0.90) and useful lactose predictions (R² = 0.88). Neither Vis/NIR reflectance nor transmittance spectroscopy lead to an acceptable prediction of the milk urea content. Transmittance spectroscopy can thus be used to predict the 3 major milk components, but with lower accuracy for fat and crude protein than the reflectance mode. Moreover, the small sample thickness (1 mm) required for NIR transmittance measurement considerably complicates its online use.     

Evaluation of Least Squares Support Vector Machine Regression and other Multivariate Calibrations in Determination of Internal Attributes of Tea Beverages

This research aimed to explore the relationship between internal attributes (pH and soluble solids content) of tea beverages and diffuse reflectance spectra. Three multivariate calibrations including least squares support vector machine regression (LSSVR), partial least squares (PLS), and radial basis function (RBF) neural network were adopted for development of internal attributes determination models. Ten kinds of tea beverages including green tea and black tea were selected for visible and near infrared reflectance (Vis/NIR) spectroscopy measurement from 325 to 1,075 nm. As regard the kernel function, least squares–support vector machine regression models were built with both linear and RBF kernel functions. Grid research and tenfold cross-validation procedures were adopted for optimization of LSSVR parameters. The generalization ability of LSSVR models were evaluated by adjusting the number of samples in the training set and testing set, and sensitive wavelengths that were closely correlated with the internal attributes were explored by analyzing the regression coefficients from linear LSSVR model. Excellent LSSVR models were built with r = 0.998, standard error of prediction (SEP) = 0.111, for pH and r = 0.997, SEP = 0.256, for soluble solids content, and it can be found that the LSSVR models outperformed the PLS and RBF neural network models with higher accuracy and lower error. Six individual sensitive wavelengths for pH were obtained, and the corresponding pH determination model was developed with r = 0.994, SEP = 0.173, based on these six wavelengths. The soluble solids content determination model was also developed with r = 0.977, SEP = 0.173, based on seven individual sensitive wavelengths. The above results proved that Vis/NIR spectroscopy could be used to measure the pH and soluble solids content in tea beverages nondestructively, and LSSVR was an effective arithmetic for multivariate calibration regression and sensitive wavelengths selection.     

A feasibility study on quantitative analysis of glucose and fructose in lotus root powder by FT-NIR spectroscopy and chemometrics

The feasibility of rapid analysis of glucose and fructose in lotus root powder by Fourier transform near-infrared (FT-NIR) spectroscopy was studied. Diffuse reflectance spectra were collected between 4000 and 12,432 cm⁻¹. Calibration models established by partial least-squares regression (PLSR), interval PLS of forward (FiPLS) and backward (BiPLS), back propagation-artificial neural networks (BP-ANN) and least squares-support vector machine (LS-SVM) were compared. The optimal models for glucose and fructose were obtained by LS-SVM with the first 10 latent variables (LVs) as input. For fructose the correlation coefficients of calibration (r_c) and prediction (r_p), the root-mean-square errors of calibration (RMSEC) and prediction (RMSEP), and the residual predictive deviation (RPD) were 0.9827, 0.9765, 0.107%, 0.115% and 4.599, respectively. For glucose the indexes were 0.9243, 0.8286, 0.543%, 0.812% and 1.785. The results indicate that NIR spectroscopy technique with LS-SVM offers effective quantitative capability for glucose and fructose in lotus root powder.     

Nondestructive Measurement of Acidity of Strawberry Using Vis/NIR Spectroscopy

Vis/Near infrared reflectance spectroscopy appears to be a rapid and convenient non-destructive technique that can measure the quality and compositional attributes of many substances. This paper assesses the ability of NIR reflectance spectroscopy to estimate the acidity of strawberry. Spectra were collected from 65 samples and data was expressed as absorbance, the logarithm of the reciprocal of reflectance (log1/R). The absorbance data was subsequently compressed using wavelet transformation. Two models to predict the acidity in strawberry were constructed. A prediction model based on wavelet transform (WT) combined with partial least squares (PLS) was found better with the r of 0.856, RMSEP of 0.026, and in the confidence lever 95%.     

Rapid determination of shoot nitrogen status in rice using near infrared reflectance spectroscopy

Near infrared reflectance (NIR) spectroscopy is a rapid, cheap, simple technique which can be used to make quantitative analyses of the concentrations of nutrients in plant tissue. The application of NIR to determine nitrogen in rice was examined. The absorbance spectrum of rice (Oryza sativa L) shoot tissue was similar to that of the temperate cereal wheat even though rice tissue has a much higher silica content. A 19-filter NIR instrument was calibrated to estimate the nitrogen content of rice shoots with between 0·8 and 3·50% N by the Kjeldahl technique. The calibration model developed used three wavelengths to account for 96% of the variation in sample Kjeldahl nitrogen concentration. This model was validated using 67 samples comprising five rice varieties grown on farms in two seasons in southern New South Wales. The standard error of prediction of the model was 0·15% N. A tissue testing service using this NIR calibration is now operational for rice crops in southern New South Wales.     

A new approach to discriminate varieties of tobacco using vis/near infrared spectra

Vis/near infrared reflectance spectroscopy appears to be a rapid and convenient non-destructive technique that can measure the quality and compositional attributes of many substances. Principal component analysis (PCA), which offered a qualitative analysis of tobacco samples, was used to analyze the clustering of tobacco samples. A new method combined wavelet transform (WT) with Artificial Neural Network (ANN) was presented to establish a discrimination model. The model regarded the compressed spectra data as the input of ANN, and 80 samples were selected randomly as calibration collection whereas the remaining 20 were being prediction collection. High correlation coefficient (r=0.999) was achieved, which was better than PCA-SRA-ANN and PLS-ANN. It indicated that WT combined with ANN is an available method for variety discrimination based on the Vis/NIR spectroscopy technology. Some sensitive wave bands were also analyzed to develop tobacco varieties discrimination apparatus through PLS models.     

基于最小二乘支持向量机的白酒酒醅成分定量分析

利用近红外光谱技术实现对白酒发酵过程中酒醅主要成分的质量控制,并进行模型优化,提高性能。采用偏最小二乘法提取的潜在变量作为最小二乘支持向量机的输入变量,先后建立了白酒酒醅中酒精度、淀粉、水分、酸度的近红外定量模型,并与经无信息变量消除法波段筛选后建立的偏最小二乘模型结果进行比较。结果表明：与偏最小二乘模型相比,4 个指标的最小二乘支持向量机定量模型的相关系数（R2）、预测均方根误差以及相对分析误差3 个评价参数均有更优表现;对未知样品进行预测时,最小二乘支持向量机模型的预测准确度明显高于偏最小二乘模型。说明最小二乘支持向量机模型的准确度、稳定性及预测性能均优于偏最小二乘法模型,为白酒酒醅的品质分析方法研究提供了新的思路。     

A feasibility study on on-line determination of rice wine composition by Vis–NIR spectroscopy and least-squares support vector machines

The feasibility of visible and near infrared (Vis–NIR) spectroscopy and least-squares support vector machines (LS-SVM) for on-line determination of rice wine composition was investigated. A circle-light fibre spectrometer system was designed to collect transreflectance spectra of rice wine samples in round brown glass bottles with the bottle sealed and the labels removed. Statistical equations were established between reference data and Vis–NIR spectra by LS-SVM. Compared to partial least squares regression (PLSR), the performance of LS-SVM was slightly better, with higher correlation coefficients for validation (r_val) of 0.915, 0.888 and 0.872, and lower root mean square error of validation (RMSEP) of 0.168 (%(V V⁻¹)), 0.146 (g L⁻¹) and 0.033 for alcohol content, titratable acidity, and pH, respectively. Based on the results, it was concluded that the Vis–NIR spectrometer system was suitable for on-line wine quality determination, and LS-SVM was a reliable multivariate method for NIR analysis.     

Monitoring water uptake in whole barley (Hordeum vulgare L.) grain during steeping using near infrared reflectance spectroscopy

The aim of this study was to evaluate the use of near infrared reflectance (NIR) spectroscopy to monitor water uptake and steeping time in whole barley samples as a rapid and easy to use technique. Whole barley grain samples were steeped in water, and subsamples were analyzed for water uptake (gravimetric method) and using NIR spectroscopy. The spectra and the analytical data were used to develop partial least squares (PLS) calibrations to predict water uptake and steeping time. Cross validation models for water uptake and steeping time gave a coefficient of determination in cross validation (R²) and the standard error of prediction (SEP) of 0.90 (SEP = 5.36 g/100 g fw) and 0.92 (SEP = 3.93 h), respectively. This study showed that NIR spectroscopy combined with PLS regression showed promise as a rapid, non-destructive method to monitor and measure water uptake and steeping time in whole barley during soaking.     

Application of visible and near infrared spectroscopy for rapid and non-invasive quantification of common adulterants in Spirulina powder

The authentication of food products is critically important in a global economy in public-health and economic terms. The specific aims of this study were to evaluate the application of full-spectrum and NIR spectroscopy and to evaluate the adoption of PLS and LS-SVM models to accomplish a rapid and non-invasive quantification of the two common adulterants, flour and mungbean powder, in Spirulina powder. The results showed that, using all treatment sets, only the LS-SVM models were adequate in predicting either adulterant under both full spectra and NIR spectra. The use of NIR spectra would allow detection of adulterants even when masked by food dyes. Design value analysis indicated that the benefits per unit cost of applying the NIR spectra to quantify adulterants in Spirulina powder significantly exceeded that of using full spectra, and that the value of employing the LS-SVM models under NIR spectra exceeded that of using the PLS models.