近红外结合Si-ELM检测食醋品质指标

为了提高近红外光谱技术检测食醋中可溶性无盐固形物含量(SSFSC)的精度和稳定性,提出采用联合区间偏最小二乘(Si-PLS)筛选光谱特征区间,再利用极限学习机(ELM)算法建立非线性回归模型,并对该方法的优越性进行系统比较;试验通过交互验证优化模型相关参数,以预测时的相关系数(Rp)和预测均方根误差(RMSEP)作为模型的评价指标。结果表明,Si-PLS结合ELM算法(Si-ELM)所建模型最佳,预测结果:Rp=0.973 9,RMSEP=1.232g/100mL。说明利用近红外光谱技术可以快速准确检测食醋中的SSF-SC,Si-ELM的应用可以适当提高该预测模型的精度。     

Simultaneous measurement of total acid content and soluble salt-free solids content in Chinese vinegar using near-infrared spectroscopy

Total acid content (TAC) and soluble salt-free solids content (SSFSC) in Chinese vinegar are 2 important indicators in the assessment of its quality. This paper shows the feasibility to determine TAC and SSFSC in Chinese vinegar by near-infrared (NIR) spectroscopy. Synergy interval partial least square (Si-PLS) algorithm was performed to calibrate the regression model. The number of PLS factors and the number of intervals were optimized simultaneously by cross-validation. The performance of the model was evaluated according to root mean square error of prediction (RMSEP) and correlation coefficient (R) in the prediction set. The optimum Si-PLS model for TAC was achieved with RMSEP = 0.264 g/100 mL and R(p) = 0.9655; the optimum Si-PLS model for SSFSC was achieved with RMSEP = 1.93 g/100 mL and R(p) = 0.9302. The overall results demonstrated that NIR spectroscopy combined with Si-PLS could be utilized to determinate TAC and SSFSC in Chinese vinegar, and NIR spectroscopy has a potential to be used in vinegar industry.     

Determination of Amino Acid Nitrogen in Soy Sauce Using Near Infrared Spectroscopy Combined with Characteristic Variables Selection and Extreme Learning Machine

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.     

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.     

Feasibility study on the use of Fourier transform near-infrared spectroscopy together with chemometrics to discriminate and quantify adulteration in cocoa beans

Fourier transform near-infrared (FT-NIR) spectroscopy combined with Support Vector Machine (SVM) and synergy interval partial least square (Si-PLS) was attempted in this study for cocoa bean authentication. SVM was used to develop an identification model to discriminate between fermented cocoa beans (FC), unfermented cocoa beans (UFC) and adulterated cocoa bean (5–40 wt/wt.% content of UFC). Si-PLS model was used to quantify the addition of UFC in FC. SVM model accurately discriminated the cocoa bean samples used. After cross-validation, the optimal identification rate was 100% in both the training set and prediction set at three principal components. For quantitative analysis, Si-PLS model was evaluated according to root mean square error of prediction (RMSEP) and coefficient of correlation in prediction (R_pred). The results revealed that Si-PLS model in this work was promising. The optimal performance of Si-PLS model showed an excellent predictive potential, RMSEP = 1.68 and R_pred = 0.98 in the prediction set. The overall results indicated that FT-NIR spectroscopy together with an appropriate multivariate algorithm could be employed for rapid identification of fermented and unfermented cocoa beans as well as the quantification of UFC down to 5% in FC for quality control management.     

Novel Prediction of Total Fat Content in Cocoa Beans by FT-NIR Spectroscopy Based on Effective Spectral Selection Multivariate Regression

Total fat content is a major quality parameter that chocolate manufactures consider when selecting cocoa beans. This paper attempted the feasibility of measuring total fat content in cocoa beans by using Fourier transform near-infrared (FT-NIR) spectroscopy based on a novel systematic study on efficient spectral variables selection multivariate regression. After the efficient spectra interval selection by synergy interval partial least squares (Si-PLS), the data were treated with support vector machine regression (SVMR) leading to synergy interval support vector machine regression (Si-SVMR). Experimental results showed that the model based on the novel Si-SVMR algorithm was superior to the others. The optimum results were assessed by root-mean-square error of prediction (RMSEP) and correlation coefficient (R _pre) in the prediction set. The performance of Si-SVMR model was RMSEP?=?0.015 and R _pre?=?0.9708. This study has demonstrated that the total fat content in cocoa beans could rapidly be predicted by FT-NIR spectroscopy and Si-SVMR technique. The novel strength and accuracy of Si-SVMR in contrast to other multivariate algorithms has been derived.     

Near-infrared spectroscopy coupled chemometric algorithms for prediction of antioxidant activity of black goji berries (<Emphasis Type="Italic">Lycium ruthenicum</Emphasis> Murr.)

In this study, near-infrared (NIR) spectroscopy coupled with partial least-squares (PLS) regression and various efficient variable selection algorithms, synergy interval-PLS (Si-PLS), backward interval PLS (Bi-PLS) and genetic algorithm-PLS (GA-PLS) were applied comparatively for the prediction of antioxidant activity in black wolfberry (BW). The eight assays were used for quantification of antioxidant content. The developed models were assessed using correlation coefficients (R²) of the calibration (Cal.) and prediction (Pre.); root mean square error of prediction, RMSEP; standard Error of Cross-Validation, RMSECV and residual predictive deviation, RPD. The performance of the built model greatly improved by the application of Si-PLS, Bi-PLS and GA-PLS compared with full spectrum PLS. The R² values determined for calibration and prediction set ranged from 0.8479 to 0.9696 and 0.8401 to 0.9638, respectively. These findings revealed that NIR spectroscopy combined with chemometric algorithms can be used for quantification of antioxidant activity in BW samples.     

Determination of Pear Internal Quality Attributes by Fourier Transform Near Infrared (FT-NIR) Spectroscopy and Multivariate Analysis

This paper attempted the feasibility to determine firmness and soluble solid content (SSC) in intact pears using Fourier transform near infrared (FT-NIR) spectroscopy coupled with multivariate analysis. Principal component analysis and independent component analysis were employed comparatively to extract latent vectors from the original spectra data. Extreme learning machine (ELM) was performed to calibrate regression model. Some parameters of ELM model were optimized according to the lowest root mean square error of cross-validation in the calibration set. Moreover, the root mean square error of prediction of the calibration model was finally corrected for making it more closed to the true prediction error due to the effect of reference measurement error existing in the pear sample attribute value on the prediction error of the model. Experimental results showed that the $ R_p^2 $ and ratio performance deviation (RPD) in the prediction set were achieved as follows: $ R_p^2 $ ?=?0.81 and RPD?=?2.28 for the firmness model when ICs?=?6 and $ R_p^2 $ ?=?0.91 and RPD?=?3.43 for the SSC model when ICs?=?5. This study demonstrates that the predictive precision of the calibration model can be effectively enhanced in measurement of firmness and SSC in intact pears by use of FT-NIR spectroscopy combined with appropriate chemometrics methods.     

基于拉曼光谱的面团冻结过程中水分分布的在线监测

采用便携式拉曼光谱仪结合低场核磁研究了面团冻结过程中水分分布的变化规律,并采用联合区间最小二乘法(Si-PLS)建立了水分分布的定量光谱分析模型。研究发现,在冻结过程中,面团的水分迁移主要发生在15～25 min之间,主要表现为深层结合水的大幅上升和弱结合水的大幅下降。Si-PLS建模结果显示水分分布变化的主要光谱区域为454.5～609.0、1 035.9～1 247.8、1 628.6～1 804. 8、2 450.2～2 593.0、2701.9～2 789.1和3 042.9～3 121.8 cm~(-1)。深层结合水、弱结合水和自由水预测模型的相关系数分别为0.881 9、0. 827 3和0. 899 3,均方根误差分别为0.14%、0.17%和0. 002 5%,拉曼光谱结合Si-PLS法建立的模型具有较好的预测能力,可以实现面团冻结过程中水分分布的在线监测。     

基于近红外光谱技术的老陈醋可溶性固形物定量分析

可溶性固形物含量(SSC)是食品行业的重要技术参数之一。利用近红外光谱技术对不同醋龄的老陈醋SSC进行分析。在不同光谱预处理下,分别采用主成分回归(PCR)和偏最小二乘法(PLS)建立SSC的定量分析模型。结果表明,采用5点平滑预处理后,利用PLS建立的老陈醋SSC的定量分析模型最优,其校正集的相关系数R为0.999 9,校正标准偏差(RMSEC)为0.038 3,预测标准偏差(RMSEP)和交叉验证标准偏差(RMSECV)分别为0.082 1,0.096 4。表明采用近红外光谱技术对不同醋龄的老陈醋SSC进行定量分析建模是可行的。     

基于高光谱成像技术的祁门红茶等级无损检测

为实现快速无损的茶叶产品等级评估,应用近红外（900~1700 nm）高光谱成像技术对6个等级的祁门红茶进行分类。首先利用线性和非线性降维方法对高光谱数据进行可视化处理,可视化算法包括线性方法的主成分分析（Principal Component Analysis,PCA）、多维尺度变换（Multi-Dimensional Scaling,MDS）,和非线性方法的t分布随机邻域嵌入（t-Distributed Stochastic Neighbour Embedding,t-SNE）、Sammon非线性映射。其次利用支持向量机（Support Vector Machine,SVM）和极限学习机（Extreme Learning Machine,ELM）建立分类模型来鉴定祁门红茶的不同等级。最后利用SVM和ELM分类模型对高光谱图像每个像素点进行识别,得到预测图。结果表明,t-SNE可以将6个等级的祁门红茶分在六个不同的簇,SVM和ELM的预测集准确率分别为100%和96.35%。t-SNE可视化效果最佳,SVM的检测模型能够有效地对祁门红茶六个等级进行识别。本文为茶叶产品等级的快速、无损检测提供了一种有效的方法,对茶叶产品的质量控制、真伪检测和掺假检测具有重要意义。     

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.     

Comparison of near-infrared and dielectric spectra for quantitative identification of bovine colostrum adulterated with mature milk

The nonhomogeneity of bovine colostrum leads to strong scattering effects for electromagnetic waves, which affects the application of electromagnetic spectroscopy in detecting colostrum. This work aimed to compare the performance of near-infrared spectroscopy (NIRS) and dielectric spectroscopy (DS) in quantitatively predicting the content of mature milk as an adulterant in colostrum. The near-infrared spectra in the range of 833 to 2,500 nm and the dielectric spectra in the range of 20 to 4,500 MHz of 150 adulterated colostrum samples containing 0 to 50% mature milk were analyzed. The different proportions of mature milk in colostrum significantly changed near-infrared and dielectric spectra. The principal component analysis score plot showed that both NIRS and DS could identify the proportion of mature milk in colostrum, but the 2 methods had different characteristics. Linear partial least squares regression and nonlinear least squares support vector machine (LSSVM) models based on near-infrared and dielectric spectra were established to identify doping proportions. The results showed that DS had better identification performance with a root-mean-square error of prediction of 4.9% and a residual prediction deviation of 3.441 by successive projection algorithm LSSVM, whereas NIRS was relatively weak with a root-mean-square error of prediction of 7.3% and a residual prediction deviation of 2.301 by full-spectra LSSVM. This work provides important insights for the quantitative prediction of nonhomogeneous liquid food by DS.     

遗传算法联合LS-SVM的苹果原醋成分定量分析

利用近红外光谱技术对苹果原醋中的重要指标进行定量分析,并进行模型优化以提高性能。采用遗传偏最小二乘法（GA-PLS）提取的特征波长作为最小二乘支持向量机（LS-SVM）的输入变量,先后建立苹果原醋中总酸、可溶性固形物的近红外定量模型,并与建立的偏最小二乘（PLS）模型结果进行比较。用决定系数（R2）、预测均方根误差（RMSEP）以及相对分析误差（RPD）对模型进行评价,确定最佳建模方法。结果表明,相比于PLS模型,总酸及可溶性固形物指标的LS-SVM定量模型的R2、RMSEP以及RPD值均有更好的表现,且在进行独立测试集验证时,LS-SVM模型的预测精度也明显优于PLS模型。说明遗传算法联合LS-SVM建立的定量模型有很高的准确度及稳定性,可以应用于苹果原醋总酸和可溶性固形物含量的快速检测。     

基于苦荞的功能醋粉对L-NNA诱导高血压大鼠降血压及抗氧化作用

制备一种基于苦荞的功能醋粉,通过给药N-硝基-L-精氨酸建立高血压大鼠模型,对该醋粉的降血压、抗氧化效果进行评估。结果表明,功能醋粉能够显著降低大鼠的血压（P＜0.05）;对于大鼠体质量、心率、肝质量的非正常增长有显著的缓解作用（P＜0.05）;能够显著地稳定大鼠血清中一氧化氮（NO）、内皮素（ET）-1及肿瘤坏死因子（TNF）-α的含量（P＜0.05）;对于大鼠肝脏的过氧化氢酶（CAT）、丙二醛（MDA）、谷胱甘肽过氧化物酶（GSH-PX）的非正常变化有极显著的缓解作用（P＜0.01）,对高血压引起的肝脏总抗氧化能力（TAC）和总超氧化物歧化酶（T-SOD）水平的降低有显著的缓解作用（P＜0.05）。说明功能醋粉能够降低大鼠血压、提升大鼠抗氧化能力,其中功能醋粉高剂量组的改善作用更为显著。     

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.     

Prediction of black tea fermentation quality indices using NIRS and nonlinear tools

Catechin content, the ratio of tea polyphenols and free amino acids (TP/FAA), as well as the ratio of theaflavins and thearubigins (TFs/TRs) are important biochemical indicators to evaluate fermentation quality. To achieve rapid determination of such biochemical indicators, synergy interval partial least square and extreme learning machine combined with an adaptive boosting algorithm, Si-ELM-AdaBoost algorithm, were used to establish quantitative analysis models between near infrared spectroscopy (NIRS) and catechin content and between TFs/TRs and TP/FAA, respectively. The results showed that prediction performance of the Si-ELM-AdaBoost mixed algorithm is superior than that of other models. The prediction results with root-mean-square error of prediction ranged from 0.006 to 0.563, the ratio performance deviation values exceeded 2.5, and predictive correlation coefficient values exceeded 0.9 in the prediction model of each biochemical indicator. NIRS combined with Si-ELM-AdaBoost mixed algorithm could be utilized for online monitoring of black tea fermentation. Meanwhile, the AdaBoost algorithm effectively improved the accuracy of the ELM model and could better approach the nonlinear continuous function.     

软枣猕猴桃玫瑰醋发酵条件响应面法优化

该试验以软枣猕猴桃、玫瑰为原料制作果醋,在单因素试验的基础上,利用响应面法对软枣猕猴桃玫瑰果醋的发酵工艺条件进行优化。采用Design-Expert 8.0.6设计软件进行分析,分析结果显示最优发酵条件为：温度32 ℃,初始pH值3.4,初始酒精度为7%vol。在此条件下果醋酸度最高可达5.39 g/100 mL。三个因素与果醋酸度之间建立的回归模型均为极显著,可用于实际生产预测。     

基于NIR技术和ELM的烤烟烟叶自动分级

为解决初烤烟叶收购中人工分级主观因素影响较大的问题,提出了一种基于近红外（NIR）光谱技术结合极限学习机（ELM）算法自动鉴别烟叶等级的方法。文章首次提出基于品质相似、价格接近原则的烟叶收购分组方法,通过交互检验优化ELM分组、分级模型的隐节点数,并与K最近邻法（KNN）、支持向量机（SVM）和随机森林（RF）等多分类算法进行了比较。结果表明:ELM分类模型参数自动优化、训练时间短、稳定性和预测能力较好,2014年（数据集A）、2015年（数据集B）烟叶收购国标样本上、中、下等烟外部预测分组正确率分别为95.77%和94.23%,数据集A和B的上、中、下等烟各组样本外部预测分级正确率分别为85.71%、86.67%、100%和100%、92.86%、92.86%。因此,采用NIR技术结合ELM能准确鉴别初烤烟叶等级,可为烤烟烟叶收购质量等级评价提供一种新技术。      

Application of successive projections algorithm for variable selection to determine organic acids of plum vinegar

Visible and near infrared (Vis/NIR) spectroscopy was investigated to determine the acetic, tartaric and lactic acids of plum vinegar based on a newly proposed combination of successive projections algorithm-least squares-support vector machine (SPA-LS-SVM). SPA, compared with regression coefficients (RC), was applied to select effective wavelengths (EWs) with least collinearity and redundancies. Five concentration levels (100%, 80%, 60%, 40% and 20%) of plum vinegar were studied. Multiple linear regression (MLR) and partial least squares (PLS) models were developed for comparison. The results indicated that SPA-LS-SVM achieved the optimal performance for three acids comparing with full-spectrum PLS, SPA-MLR, SPA-PLS, RC-PLS and RC-LS-SVM. The root mean square error of prediction (RMSEP) was 0.3581, 0.0714 and 0.0201 for acetic, tartaric and lactic acids, respectively. The overall results indicated that Vis/NIR spectroscopy incorporated to SPA-LS-SVM could be applied as an alternative fast and accurate method for the determination of organic acids of plum vinegars.