Variable Selection in Visible and Near-Infrared Spectral Analysis for Noninvasive Determination of Soluble Solids Content of ‘Ya’ Pear

Informative variable selection or wavelength selection plays an important role in the quantitative analysis of near-infrared (NIR) spectra because the modern spectroscopy instrumentations usually have a high resolution and the obtained spectral data sets may have thousands of variables and hundreds or thousands of samples. In this study, a new combination of Monte Carlo–uninformative variable elimination (MC-UVE) and successive projections algorithm (SPA; MC-UVE-SPA) was proposed to select the most effective variables. MC-UVE was firstly used to eliminate the uninformative variables in the raw spectra data. Then, SPA was applied to determine the variables with the least collinearity. A case study was done based on the NIR spectroscopy for the non-destructive determination of soluble solids content (SSC) in ‘Ya’ pear. A total of 160 samples were prepared for the calibration (n?=?120) and prediction (n?=?40) sets. Three calibration algorithms including linear regressions of partial least square regression (PLS) and multiple linear regression (MLR), and nonlinear regression of least-square support vector machine (LS-SVM) were used for model establishment by using the selected variables by SPA, UVE, MC-UVE, UVE-SPA, and MC-UVE-SPA, respectively. The results indicated that linear models such as PLS and MLR were more effective than nonlinear model such as LS-SVM in the prediction of SSC of ‘Ya’ pear. In terms of linear models, different variable selection methods can obtain a similar result with the RMSEP values range from 0.2437 to 0.2830. However, combination of MC-UVE and SPA was helpful for obtaining a more parsimonious and efficient model for predicting the SSC values in ‘Ya’ pear. Twenty-two effective variables selected by MC-UVE-SPA achieved the optimal linear MC-UVE-SPA-MLR model compared with other all developed models by balancing between model accuracy and model complexity. The coefficients of determination (r ²), root mean square error of prediction, and residual predictive deviation by MC-UVE-SPA-MLR were 0.9271, 0.2522, and 3.7037, respectively.     

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.     

Nondestructive measurement of soluble solid content of navel orange fruit by visible-NIR spectrometric technique with PLSR and PCA-BPNN

A relationship was established between the soluble solid content (SSC) of navel orange fruit determined by destructive measurement and visible-near infrared diffuse reflectance spectra in the wavelength range of 350-1800 nm. Multiplicative scatter correction (MSC) and standard normal variate correction (SNV) were applied to the spectra, partial least squares regression (PLSR) and back propagation neural network (BPNN) based on principal component analysis (PCA) were used to develop the models for predicting the SSC of intact navel orange fruit. Thirty-eight unknown samples were used to evaluate the performance of these models. The principal component analysis-back propagation (PCA-BPNN) method with MSC spectral pretreatment obtain the best predictive results, resulting in correlation coefficient, root mean square error of prediction (RMSEP), average difference between predicted and measured values (Bias) of 0.90, 0.68 °Brix and 0.16 °Brix, respectively. Experimental results indicate that PCA-BPNN is a suitable tool to model the non-linear complex system, with additional advantages over PLSR, and the vis/NIR spectrometric technique can be used for measuring the SSC of intact navel orange fruit, nondestructively.     

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.     

Nondestructive determination of soluble solids content and pH in tomato juice using NIR transmittance spectroscopy

The potential of near-infrared (NIR) transmittance spectroscopy to nondestructively detect soluble solids content (SSC) and pH in tomato juices was investigated. A total of 200 tomato juice samples were used for NIR spectroscopy analysis at 800–2400 nm using an FT-NIR spectrometer. Multiplicative signal correction (MSC), and the first and second derivative were applied for pre-processing spectral data. The relationship between SSC, pH, and FT-NIR spectra of tomato juice were analyzed via partial least-squares (PLS) regression. PLS regression models were able to predict SSC and pH in tomato juices. The r _c, RMSEC, RMSEP, and RMSECV for SSC were 0.92, 0.0703°Brix, 0.150°Brix, and 0.138°Brix, respectively, whereas those values for pH were 0.90, 0.0333, 0.0316, and 0.0489, respectively. It is concluded that the combination of NIR transmittance spectroscopy and PLS methods can be used to provide a technique of convenient, versatile, and rapid analysis for SSC and pH in tomato juices.     

京郊鲜食杏白利糖度的便携式光谱快速无损检测方法研究

目的 建立京郊鲜食杏白利糖度的定量分析预测模型, 实现对京郊鲜食杏品质的快速无损检测。方法 使用便携式近红外光谱仪采集900~1700 nm下鲜食杏的漫反射光谱信息, 使用多元散射校正(multiplicative scatter correction, MSC)、标准正态变量变换(standard normal variable transformation, SNV)和Savitzky-Golay卷积平滑(Ssavitzky – Ggolay smooth, S-G)对原始光谱数据进行预处理, 使用Kennard-Stone (K-S)算法以3:1比例将样本集划分成校正集和预测集, 利用竞争自适应重加权采样(competitive adaptive reweighted sampling, CARS)算法和连续投影算法(successive projections algorithm, SPA)对光谱进行特征波长筛选, 使用偏最小二乘回归(partial least squares regression, PLSR)算法建立京郊鲜食杏白利糖度的预测模型。结果 以MSC+S-G+ CARS+PLSR算法建立的北京鲜食杏的白利糖度预测模型取得较好的预测精度, 模型的校正集均方根误差、校正集相关系数、预测集均方根误差、预测集相关系数分别为0.3502、0.9747、0.4698、0.9616。结论 基于便携式近红外光谱技术所采集数据构建的京郊鲜食杏白利糖度预测模型准确性较高, 可以快速准确检测鲜食杏白利糖度, 从而实现对鲜食杏品质的快速无损检测, 为鲜食杏的品质检测提供了理论依据和方法指导。     

Application of Portable and Handheld Infrared Spectrometers for Determination of Sucrose Levels in Infant Cereals

Sucrose coating of breakfast cereals is used to enhance the flavor and attractiveness of the final product but there is a need for monitoring its levels to meet consumer health concerns associated with sugar consumption. Our objective was to evaluate the use of portable (mid-infrared, MIR) and handheld (near-infrared, NIR) systems for rapid, simple and reliable determination of sucrose content in breakfast cereal products. Cereal-based and sucrose-coated samples were provided by an Ohio snack food company. Samples were ground and spectra were collected using portable ATR-MIR (Cary 630) and handheld NIR (microPHAZIR) spectrometers. Reference sucrose levels were determined by high-performance liquid chromatography (HPLC). Partial least squares regression (PLSR) was used to develop calibration regression models for prediction of sucrose levels in breakfast cereals based on spectral data. Sucrose levels in uncoated (n?=?28) and coated (n?=?62) cereal samples were on average of 1.2?±?0.7 and 11.8?±?3.5 g/100 g, respectively. Similar calibration (n?=?85) model performances were obtained for determination of sucrose content by using the portable MIR and handheld NIR instruments with standard error of cross-validation (SECV) of 1.45 %. However, superior predictive ability was obtained with the portable MIR unit using a validation set (n?=?20, SEP?=?1.27 % and RPD?=?4.41). Regression models using NIR spectrum of the cereal through a polyethylene bag resulted in reduction of the model goodness of fit and RPD values. Results support the application of handheld NIR and portable MIR spectrometers for close-to-real-time analysis of sucrose levels in breakfast cereals providing simple, rapid and reliable prediction for quality assurance.     

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.     

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.     

Application of Competitive Adaptive Reweighted Sampling Method to Determine Effective Wavelengths for Prediction of Total Acid of Vinegar

In this paper, near-infrared (NIR) spectroscopy coupled with wavelength selection methods was used to predict total acid of vinegar. Three wavelength selection methods including competitive adaptive reweighted sampling (CARS), Monte Carlo uninformative variable elimination (MC-UVE), and moving window partial least squares (MWPLS) were employed to select the key wavelengths. Five wavelengths including 4,348, 4,694, 5,365, 7,104, and 7,236 cm⁻¹ were selected by CARS method. Least squares (LS) regression model was built on the selected wavelengths. Compared to the partial least squares regression models based on full spectrum and wavelengths selected by MC-UVE and MWPLS, the performance of LS model was better, with higher determination coefficient for test (r ²) of 0.997, and lower root mean square error of prediction of 0.13 g/100 ml. Based on the results, it was concluded that NIR spectroscopy combined with CARS methods seem to be a rapid and effective alternative to the classical methods for the prediction of total acid of vinegar.     

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.