Predicting acetic acid content in the final beer using neural networks and support vector machine

Most acetic acid found in beer is produced by yeast during fermentation. It contributes significantly to beer taste, especially when its content is higher than the taste threshold in beer. Therefore, the control of its content is very important to maintain consistent beer quality. In this study, artificial neural networks and support vector machine (SVM) were applied to predict acetic acid content at the end of a commercial‐scale beer fermentation. Relationships between beer fermentation process parameters and the acetic acid level in the fermented wort (beer) were modelled by partial least squares (PLS) regression, back‐propagation neural network (BP‐NN), radial basis function neural network (RBF‐NN) and least squares‐support vector machine (LS‐SVM). The data used in this study were collected from 146 production batches of the same beer brand. For predicting acetic acid content, LS‐SVM and RBF‐NN were found to be better than BP‐NN and PLS. For the comparison of RBF‐NN and LS‐SVM, RBF‐NN had a better reliability of model, but lower reliability of prediction. SVM had better generalization, but lower reliability of model. In summary, LS‐SVM was better than RBF‐NN modelling for the prediction of acetic acid content during the commercial beer fermentation in this study. Copyright © 2013 The Institute of Brewing & Distilling     

Predicting cow milk quality traits from routinely available milk spectra using statistical machine learning methods

Numerous statistical machine learning methods suitable for application to highly correlated features, as those that exist for spectral data, could potentially improve prediction performance over the commonly used partial least squares approach. Milk samples from 622 individual cows with known detailed protein composition and technological trait data accompanied by mid-infrared spectra were available to assess the predictive ability of different regression and classification algorithms. The regression-based approaches were partial least squares regression (PLSR), ridge regression (RR), least absolute shrinkage and selection operator (LASSO), elastic net, principal component regression, projection pursuit regression, spike and slab regression, random forests, boosting decision trees, neural networks (NN), and a post-hoc approach of model averaging (MA). Several classification methods (i.e., partial least squares discriminant analysis (PLSDA), random forests, boosting decision trees, and support vector machines (SVM)) were also used after stratifying the traits of interest into categories. In the regression analyses, MA was the best prediction method for 6 of the 14 traits investigated [curd firmness at 60 min, α_S1-casein (CN), α_S2-CN, κ-CN, α-lactalbumin, and β-lactoglobulin B], whereas NN and RR were the best algorithms for 3 traits each (rennet coagulation time, curd-firming time, and heat stability, and curd firmness at 30 min, β-CN, and β-lactoglobulin A, respectively), PLSR was best for pH, and LASSO was best for CN micelle size. When traits were divided into 2 classes, SVM had the greatest accuracy for the majority of the traits investigated. Although the well-established PLSR-based method performed competitively, the application of statistical machine learning methods for regression analyses reduced the root mean square error compared with PLSR from between 0.18% (κ-CN) to 3.67% (heat stability). The use of modern statistical machine learning methods for trait prediction from mid-infrared spectroscopy may improve the prediction accuracy for some traits.     

Recent Advances in Wavelength Selection Techniques for Hyperspectral Image Processing in the Food Industry

During the past decade, hyperspectral imaging (HSI) has been rapidly developing and widely applied in the food industry by virtue of the use of chemometric techniques in which wavelength selection methods play an important role. This paper is a review of such variable selection methods and their limitations, describing the basic taxonomy of the methods and their respective advantages and disadvantages. Special attention is paid to recent developments in wavelength selection techniques for HSI in the field of food quality and safety evaluations. Typical and commonly used methods in HSI, such as partial least squares regression, stepwise regression and spectrum analysis, are described in detail. Some sophisticated methods, such as successive projections algorithm, uninformative variable elimination, simulated annealing, artificial neural network and genetic algorithm methods, are also discussed. Finally, new methods not currently used but that could have substantial impact on the field are presented. In short, this review provides an overview of wavelength selection methods in food-related areas and offers a thoughtful perspective on future potentials and challenges in the development of HSI systems.     

Determination of Tert‐Butylhydroquinone in Vegetable Oils Using Surface‐Enhanced Raman Spectroscopy

Monitoring the level of tert‐butylhydroquinone (TBHQ), a permissible antioxidant additive in edible vegetable oils in many countries, is important to ensure that oils and products that contain them comply with the relevant import regulations. Surface‐enhanced Raman spectroscopy (SERS) technology coupled with chemometric methods including partial least squares (PLS) and support vector machine (SVM) regression was applied to determine levels of TBHQ in spiked corn oils (0 to 500 mg/kg, n = 40) and commercial vegetable oils (0 to 99.7 mg/kg, n = 25). The lowest detectable concentration was 5 mg/L for TBHQ in standard solutions and 10 mg/kg of TBHQ in vegetable oils from various plant sources. The TBHQ levels predicted by the PLS or SVM model had a high correlation with actual TBHQ levels in commercial oil samples (SVM: R² = 0.972, ratio of performance to deviation [RPD] = 5.55, root mean square error [RMSE] = 5.73 mg/kg; PLS: R² = 0.976, RPD = 6.43, RMSE = 4.94 mg/kg), indicating great potential of SERS methods for detection and quantification of TBHQ in oils from a variety of sources.     

基于微型近红外光谱仪的油菜籽含油率模型参数优化研究

为实现油菜籽含油率快速无损检测,采用微型近红外光谱仪,结合竞争性自适应重加权(CARS)、遗传算法(GA)、连续投影算法(SPA)、无信息变量消除法(UVE)、向后区间偏最小二乘法(BIPLS)、联合区间偏最小二乘法(SIPLS)等方法优选油菜籽含油率近红外光谱特征波长,建立偏最小二乘回归(PLSR)和最小二乘支持向量机(LS-SVM)定量分析模型,同时对LS-SVM模型参数进行优化。研究表明,对PLSR模型,BIPLS+GA优选的26个特征波长建模效果最好,其预测相关系数(Rp)和预测均方根误差(RMSEP)分别为0.9330和0.0075,对LS-SVM模型,SIPLS+GA优选的13个特征波长建模效果最好,预测相关系数(Rp)和预测均方根误差(RMSEP)分别0.9192和0.0055。证明了波长优选和参数优化可有效简化油菜籽含油率近红外光谱定量分析模型,提高模型预测精度和稳定性,为进一步拓展微型近红外光谱仪的应用提供技术参考。     

Development of a multispectral imaging system for online quality assessment of pomegranate fruit

The objective of this study was to develop a prototype multispectral imaging system for online quality assessment on pomegranate fruit. At first, a visible/near infrared spectroscopy (400–1100 nm) was tested for non-destructive determination of total soluble solids, titratable acidity, and pH. The spectral data were analyzed using the partial least square analysis. Then to establish consistent multispectral imaging system, the highest absolute values of β-coeffcients correspond to wavelengths from the best partial least square calibration model were selected and used for identifying the optimal wavelengths. Consequently, a multispectral imaging system was developed based on the effective wavelengths 700, 800, 900, and 1000 nm. The performance of the developed multispectral imaging system was evaluated by multiple linear regression models. The multiple linear regression model predict total soluble solids with r = 0.97, root mean square error of calibration = 0.21°Brix, and ratio performance deviation = 6.7 °Brix. Also, the results showed that the models had good predictive ability for pH and titratable acidity. Results showed that the developed multispectral imaging system based on the optimal wavelengths could be used for online quality assessment of pomegranate fruit.     

Determination of Postharvest Quality of Cucumbers Using Nuclear Magnetic Resonance and Electronic Nose Combined with Chemometric Methods

The purpose of this study was to investigate changes in water status and flavor characteristics of cucumbers during postharvest storage and to trace the quality attributes using partial least squares (PLS) and support vector machine (SVM). The results showed that four distinct water populations were identified in cucumbers by nuclear magnetic resonance (NMR), and the changes of water mobility and distribution occurred mainly in pulp of cucumbers. Flavor characteristics of cucumbers at different storage stages were distinguished by electronic nose (e-nose), and four clusters could be achieved through hierarchical clustering analysis. Comparison of two models, excellent prediction performances for firmness, pH, SSC, and ΔE of postharvest cucumbers were obtained using a combination method of e-nose technology and SVM algorithm. This study indicated that there were significant changes in the quality parameters of cucumbers during postharvest storage, which were related to water status and flavor characteristics. The combination of e-nose technology with the SVM algorithm offers a promising technique to monitor cucumber quality.     

Predicting beef tenderness using color and multispectral image texture features

The objective of this study was to investigate the usefulness of raw meat surface characteristics (texture) in predicting cooked beef tenderness. Color and multispectral texture features, including 4 different wavelengths and 217 image texture features, were extracted from 2 laboratory-based multispectral camera imaging systems. Steaks were segregated into tough and tender classification groups based on Warner-Bratzler shear force. The texture features were submitted to STEPWISE multiple regression and support vector machine (SVM) analyses to establish prediction models for beef tenderness. A subsample (80%) of tender or tough classified steaks were used to train models which were then validated on the remaining (20%) test steaks. For color images, the SVM model correctly identified tender steaks with 100% accurately while the STEPWISE equation identified 94.9% of the tender steaks correctly. For multispectral images, the SVM model predicted 91% and STEPWISE predicted 87% average accuracy of beef tender.     

Predicting beef tenderness using color and multispectral image texture features

The objective of this study was to investigate the usefulness of raw meat surface characteristics (texture) in predicting cooked beef tenderness. Color and multispectral texture features, including 4 different wavelengths and 217 image texture features, were extracted from 2 laboratory-based multispectral camera imaging systems. Steaks were segregated into tough and tender classification groups based on Warner–Bratzler shear force. The texture features were submitted to STEPWISE multiple regression and support vector machine (SVM) analyses to establish prediction models for beef tenderness. A subsample (80%) of tender or tough classified steaks were used to train models which were then validated on the remaining (20%) test steaks. For color images, the SVM model correctly identified tender steaks with 100% accurately while the STEPWISE equation identified 94.9% of the tender steaks correctly. For multispectral images, the SVM model predicted 91% and STEPWISE predicted 87% average accuracy of beef tender.     

玉米籽粒霉变等级高光谱图像检测方法研究

为了快速、无损检测出储藏玉米籽粒不同霉变状况,提升玉米收储环节质检效率,尝试利用高光谱成像技术结合机器学习算法构建玉米籽粒霉变等级分类模型。采集400～1 000 nm波段范围内玉米籽粒高光谱图像,以测定的真菌孢子数为依据,将籽粒霉变状态划分为健康、轻度霉变、中度霉变和重度霉变4个等级,采用随机蛙跳(RF)算法优选出7个光谱特征变量,针对特征波段图像,利用Tamura算法共提取出21个纹理特征变量,基于颜色矩阵提取出21个颜色特征变量。进一步结合支持向量机(SVM)、极限学习机(ELM)和偏最小二乘回归(PLSR)3种算法分别建立基于光谱、图像和图谱特征融合的玉米籽粒霉变等级分类模型。经分析比较,融合光谱和图像特征并结合ELM算法建立的分类模型用于玉米籽粒霉变等级识别效果最优,训练集和测试集分类准确率(Acc)分别为94.21%和93.86%,并将玉米籽粒霉变等级进行可视化表达。     

Near‐infrared spectroscopy and data analysis for predicting milk powder quality attributes

Near‐infrared (NIR) spectroscopy is a rapid analytical method for food products. In this study, NIR spectroscopy, data pretreatment techniques and multivariate data analysis were used to predict fine particle size fraction, dispersibility and bulk density of various milk powder samples, which are believed to have a significant impact on milk powder quality. Predictive models using partial least‐squares (PLS) regression were developed using NIR spectra and milk powder physical and functional properties, and it was concluded that the PLS models predicted milk powder quality with an accuracy of 88‐90 per cent.     

基于高光谱成像的马铃薯环腐病无损检测

为探讨高光谱成像技术无损检测马铃薯环腐病的可行性,采用反射高光谱（980～1 650 nm）成像技术,以120 个马铃薯样本（合格60 个,环腐60 个）为研究对象,对比多元散射校正、标准正态变换、卷积＋一阶导数等对建模的影响,优选出多元散射校正的光谱预处理方法;然后基于偏最小二乘回归系数法提取9 个特征波长（993、1 005、1 009、1 031、1 112、1 162、1 165、1 225、1 636 nm）,建立特征波长下马铃薯环腐病的2 类线性判别分析（linear discriminant analysis,LDA）模型和4 类支持向量机（support vector machine,SVM）模型,即Fisher-LDA、马氏距离-LDA、线性核SVM、径向基核SVM、多项式核SVM和S型核SVM。结果表明,LDA模型中马氏距离法最优,SVM模型中S型核SVM最优,LDA模型整体优于SVM模型,最终确定基于马氏距离LDA的马铃薯环腐病判别模型为最佳模型,校正集、验证集识别率分别为100%和93.33%。实验结果表明高光谱无损检测马铃薯环腐病具有可行性。     

太赫兹时域光谱识别4种食用油真实性的研究

采用太赫兹时域光谱系统(THz-TDS),研究了4种食用油(黑芝麻油、芝麻油、小磨香油和花生油)在0.2~1.6 THz波段的延时特性和折射率特性。使用主成分分析法(PCA),根据累计贡献率的大小提取光谱的特征数据。提取了4个主成分(累计贡献率大于95%)作为一个支持向量机(SVM)模型的输入用于识别食用油的种类。结果表明:结合主成分分析法,通过选择合适的支持向量机核函数及其参数,食用油种类识别的正确率可达到93%;通过与主成分回归(PCR)、偏最小二乘回归(PLS)和后向(BP)神经网络方法的比较,支持向量机结合主成分分析(PCA-SVM)方法具有更突出的分类性能,同时也说明了采用太赫兹时域光谱,结合化学计量学方法精准鉴别食用油种类的可行性。     

高光谱图像对白萝卜糠心的无损检测

为实现白萝卜异常品质糠心的无损检测,构建高光谱图像技术检测白萝卜糠心的检测系统。获取了光源透射、反射和半透射模式下白萝卜的高光谱图像信息,结合偏最小二乘分析（partial least squares discriminantanalysis,PLS-DA）、支持向量机（support vector machine,SVM）、人工神经网络（artificial neural network,ANN）3 种算法分别建立白萝卜糠心的识别模型。结果表明：3 种检测模式中,基于透射模式的高光谱图像系统检测准确率最高;3 种预测模型中,ANN模型优于PLS-DA和SVM模型。其中,基于透射模式的ANN模型,高光谱图像对萝卜糠心的检测总体准确率达94.3%,效果最好。因此,采用透射模式的高光谱图像技术对白萝卜糠心的检测是可行的。     

Application of near infrared spectroscopy for the rapid determination of antioxidant activity of bamboo leaf extract

This study was carried out to evaluate the feasibility of using near infrared (NIR) spectroscopy for determining three antioxidant activity indices of the extract of bamboo leaves (EBL), specifically 2,2-diphenyl-1-picrylhydrazyl (DPPH), ferric reducing/antioxidant power (FRAP), and 2,2′-azinobis-(3-ethylbenz-thiazoline-6-sulfonic acid) (ABTS). Four different linear and nonlinear regressions tools (i.e. partial least squares (PLS), multiple linear regression (MLR), back-propagation artificial neural network (BP-ANN), and least squares support vector machine (LS-SVM)) were systemically studied and compared in developing the model. Variable selection was first time considered in applying the NIR spectroscopic technique for the determination of antioxidant activity of food or agricultural products. On the basis of these selected optimum wavelengths, the established MLR calibration models provided the coefficients of correlation with a prediction (r_pre) of 0.863, 0.910, and 0.966 for DPPH, FARP, and ABTS determinations, respectively. The overall results of this study revealed the potential for use of NIR spectroscopy as an objective and non-destructive method to inspect the antioxidant activity of EBL.     

高光谱成像技术预测香葱贮藏品质

香葱是一种保质期很短的重要调味食品,水分与叶绿素是评估香葱采后品质的重要指标。本文旨在使用无损检测技术获取香葱在采后不同存储条件下的水分及叶绿素分布情况。实验采用高光谱成像技术获取431~962 nm波段的香葱反射光谱数据,通过卷积平滑（SG）、多元散射校正（MSC）、标准正态变异（SNV）三种预处理方法对原始光谱进行相应转换,并分别建立水分和叶绿素含量预测模型,比较模型预测精度后,选用降噪效果最好的MSC作为光谱预处理方法。随后使用竞争自适应加权采样算法分别选出11个和20个特征波段用于水分与叶绿素含量的预测。基于优选特征波段,利用偏最小二乘回归算法和支持向量机回归算法建立水分和叶绿素含量的预测模型。所建水分与叶绿素含量预测模型的最高预测决定系数分别达到0.9046和0.9143。最后根据所建模型取得不同存储条件下香葱水分及叶绿素含量分布图。综上,高光谱成像技术可用于快速无损检测香葱水分及叶绿素分布情况。本研究为后续便携式果蔬水分及叶绿素分布检测仪器的开发提供了理论依据。     

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.     

基于高光谱技术检测小麦粉灰分含量

目的 基于高光谱技术实现对小麦粉灰分含量的准确检测。方法 利用高光谱成像技术采集小麦粉的光谱数据,建立基于偏最小二乘法(partial least squares regression,PLSR)和深度极限学习机(deep extreme learning machines,DELM)的小麦粉灰分含量预测模型;通过分析3种预处理算法和4种波长选择算法,分别选出最佳的预处理与波长选择方法,最后构建基于特征波段光谱信息的预测模型,并对结果进行比较。结果 标准正态变量校正(standard normal variable,SNV)为最佳预处理方法;连续投影算法(successive projections algorithm,SPA)相较于随机森林(random forest,RF)、无信息变量消除(uninformative variable elimination,UVE)和遗传算法(genetic algorithm,GA)选择特征波长的模型更优;DELM模型更适用于灰分含量的检测,最优模型的测试集决定系数为0.968,预测集均方根误差为0.024。结论 高光谱成像技术可以快速、精准的...     

基于高光谱成像技术的水果表面农药残留无损检测

在高光谱成像技术的基础上,提出了一种应用于水果表面农药残留的无损检测方法。对采集数据进行预处理和特征提取,通过细菌群体趋药性算法找到最优的最小二乘支持向量机参数,建立农残检测模型,并与最小二乘支持向量机模型进行比较,验证该模型的优越性和准确性。结果表明,基于连续投影法特征波长结合文中检测模型具有最高的检测精度,其准确率达97.92%。     

基于高光谱成像技术快速检测鸡蛋中二十二碳六烯酸与虾青素含量

目的：本研究利用高光谱成像技术结合机器学习研发一种快速检测鸡蛋中DHA与虾青素含量的技术。方法：利用高光谱成像仪采集全蛋、去壳鸡蛋和蛋黄在400-1000nm波长下的光谱数据,并使用高效液相色谱及气相色谱测定鸡蛋的DHA与虾青素含量。将样本集划分为训练集和预测集,分别采用Savitzky-Golay求导法、傅里叶变换法及小波变换法对原始光谱进行降噪处理。通过遗传算法对原始光谱及降噪后的光谱提取特征波长,分别建立特征波长与全蛋、去壳鸡蛋和蛋黄中DHA、虾青素的偏最小二乘法、支持向量机、bp人工神经网络预测模型。结果：在预测鸡蛋中DHA含量模型中,基于蛋黄特征光谱的模型预测能力最强。其中,一阶导数的差分步长为5的偏最小二乘法模型预测效果最好,其训练集、预测集的决定系数分别为0.999与0.985。在预测鸡蛋中虾青素含量的模型中,基于蛋黄特征光谱的预测能力最强。其中,二阶导数的差分步长为8的支持向量机模型预测效果最好,其中训练集、预测集的决定系数分别为0.942与0.960。结论：利用高光谱成像技术, 可以实现蛋黄中DHA和AST的快速检测。