Compositional Analysis of Powdered Cocoa Products by Near Infrared Reflectance Spectroscopy

Near infrared reflectance (NIR) spectroscopy was used to measure moisture, fat, and sucrose in powdered cocoa products. Spectra of a series of known samples were recorded and multiple linear regression techniques were used to relate the concentrations of each parameter to reflectance measurements at selected wavelengths. Precision and accuracy were estimated to evaluate the potential application of the NIR spectroscopy in the quality control of powdered cocoa products. Results showed that moisture, fat, and sucrose could be analyzed in powdered cocoa products by near infrared reflectance spectroscopy because good correlation coefficients and low standard errors were achieved in prediction study.     

Quality Assessment of Poultry Egg Based on Visible–Near Infrared Spectroscopy and Radial Basis Function Networks

In this research the possibility of a non-destructive prediction of two main quality parameters of poultry egg using principle component analysis and radial basis function network by the transmission visible–near infrared spectroscopy method was investigated. The studied parameters include Haugh unit and air cell height as a function of a 5-week storage duration at room (25°C and 40% relative humidity) and refrigerator (5°C and 75% relative humidity) conditions. The spectra were interpreted and a radial basis function network model was developed for both storage conditions at wavelength ranges of 300–1100 nm. The developed models yielded a good prediction accuracy of Haugh unit for intact egg (R² value 0.745 and 0.76) as well as air cell height (R² value 0.835 and 0.844) for room and refrigerator conditions, respectively. Results of the experiment showed the developed model can be used in the prediction of egg freshness indices satisfactorily.     

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.     

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.     

基于近红外光谱技术与BP-ANN算法的豆粕品质快速检测

应用近红外漫反射光谱技术结合误差反向传递人工神经网络(BP-ANN)算法,建立豆粕品质(包括水分、粗蛋白、残油)的定量分析模型。将豆粕漫反射吸收光谱数据进行SNV、DT、SG求导、SG平滑和均值中心化处理,然后采用偏最小二乘方法(PLS)降维获取主成分,并优化选择合适的隐含层节点数、隐含层和输出层转化函数,建立校正模型,并用验证样品对校正模型进行验证。结果显示,BP-ANN法建立的水分、粗蛋白和残油的预测相关系数(R)分别为0.981、0.988、0.982,预测标准偏差(SEP)分别为0.120、0.216、0.036,均优于PLS建模方法结果,且满足传统分析方法的重复性要求,表明BP-ANN方法可用于生产过程豆粕品质的快速监控。     

基于近红外光谱技术的食用油酸值和过氧化值定量分析研究

目的 建立基于傅里叶近红外光谱技术的定量分析模型,实现快速测定食用油中酸值和过氧化值含量,保证食用油的品质安全以及跟踪食用油储藏期间的品质变化。方法 首先采用傅里叶近红外光谱仪采集食用油样品漫反射光谱,接着采用归一化(Normalize)和标准正态变换(standard normal variate,SNV)对光谱数据进行预处理,降低原始光谱中噪声的影响;其次通过随机森林(random forest,RF)和引导软收缩(bootstrapping soft shrinkage,BOSS)算法提取特征波长;最后结合径向基函数(radial basis function,RBF)神经网络和极限学习机(extreme learning machine,ELM)建立食用油酸值和过氧化值的预测模型,并与全波段的模型进行对比分析。结果 经过BOSS算法所提取的特征波段建立的模型预测效果优于RF算法以及全波段模型,酸值模型的决定系数(determination coefficient,R²)达到0.98,均方根误差(root mean square error,RMSE)达到0...     

Near-Infrared Spectroscopy for Classification of Oranges and Prediction of the Sugar Content

A nondestructive method for the classification of orange samples according to their growing conditions and geographic areas was developed using Vis/Near infrared spectroscopy. The results showed that the NIR spectra of the samples were moderately clustered in the principle component space and pattern recognition wavelet transform (WT) combined artificial neural network (BP-ANN) provided satisfactory classification results. Additionally, a partial least square (PLS) method was constructed to predict the sugar content of certain oranges. It showed excellent predictions of the sugar content of oranges, with standard error of prediction (SEP) values of 0.290 and 0.301 for Shatangju and Huangyanbendizao, respectively.     

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.     

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%.     

基于近红外化学成分特征吸收光谱的野生台蘑鉴别模型研究

本文建出一种应用近红外光谱技术鉴别野生台蘑的新方法。使用FieldSpec3便携式近红外光谱仪对包括野生台蘑在内的13种蘑菇进行漫反射光谱采集。将采集的数据经小波去噪后,对可见与近红外光谱(350～2500nm)进行峰谷筛选,所得峰谷集经主成分分析降维,取方差贡献率大于99.9%的5个主成分作为BP神经网络的输入值,建立数学模型。该模型在偏差±0.05内,对未知样本正确识别率为100%。本结果表明利用近红外漫反射光谱可以很好地鉴别野生台蘑。     

Application of near infrared reflectance spectroscopy to protein analysis of grassland herbage samples

Near infrared reflectance spectroscopy (NIRS) was evaluated for the prediction of the protein content in samples of grassland herbage taken at different stages of maturity (flowering to fruiting stage) in ‘Dehesa’ zones of Central-Western Spain. A Technicon Infra Alyzer model 500 scanning monochromator interfaced with a Hewlett-Packard 1000 minicomputer was used for the study. Protein content was predicted with NIRS data treated as log I/R using six or seven wavelengths. Calibrations were evaluated by comparing Kjeldahl analyses with those predicted by NIRS. The prediction of protein was found to be acceptable, the standard error carying between 0.56 and 0.68% in a range of protein content from 6.76 to 13.98%.     

基于傅里叶近红外光谱和电子鼻技术的苹果霉心病无损检测

为研究傅里叶近红外光谱技术(Fourier transform near infrared spectroscopy,FT-NIRS)和电子鼻技术分别结合化学计量学方法对苹果霉心病的判别效果,以“红富士”霉心病苹果和健康苹果为试材,利用近红外光谱技术,基于主成分分析建立Fisher判别和多层感知器(multi-layer perceptron,MLP)神经网络模型;同时利用电子鼻技术分别结合Fisher判别、MLP神经网络和径向基函数神经网络3种化学计量学的方法建立判别模型。根据建模集和验证集的预测准确率综合考虑,基于主成分分析建立的MLP神经网络模型和电子鼻结合MLP神经网络模型对苹果霉心病的判别效果最好,验证集中的正确判别率分别达到87.7%和86.2%。说明电子鼻和近红外光谱技术均可以较好地判别苹果霉心病。     

Quantification of total polysaccharides and triterpenoids in Ganoderma lucidum and Ganoderma atrum by near infrared spectroscopy and chemometrics

Two chemometrics, the partial least-squares (PLS) and radial basis function (RBF) network were performed to develop a quantification method for total polysaccharides and triterpenoids in Ganoderma lucidum and Ganoderma atrum from different origins based on near infrared reflectance spectroscopy (NIR). The influences of spectral window and spectral pre-treatments were initially studied in the construction of PLS model. The best result was obtained when the standard normal transformation (SNV) +1st derivative spectrum over 4100–7750 cm⁻¹ was used for the modelling. Then based on each principle, both of the two models were optimised respectively. The final results with high determination coefficient (R²) (higher than 0.973, 0.989 for PLS and RBF, respectively) and low root mean square errors of prediction (RMSEP) (low to 0.1109 and 0.01298 for polysaccharides and triterpenoids, respectively) confirm the good predictability of the two models. The overall results show that NIR spectroscopy combined with chemometrics can be efficiently utilised for accurate analysis of routine chemical compositions in G. lucidum and G. atrum.     

Assessment of maximum cooking temperatures in previously heat treated beef. Part 1: Near infrared spectroscopy

Near infrared diffuse reflectance (NIR) and transmittance (NIT) spectroscopy were studied as potential methods for determination of the previous heat treatment of beef. Ninety-four samples of M longissimus dorsi from 33 bulls were heat treated at nine different temperatures between 50 and 85°C, and analysed by NIR and NIT spectroscopy. The samples were analysed by NIR and NIT in both ‘wet’ and freeze dried states. The NIR and NIT methods were able to determine the maximum temperature of previously heat treated beef with a prediction error of 2.0-2.1°C in the temperature range 50–85°C. Freeze drying of the samples prior to analysis reduced the prediction error to 1.4°C for the NIR method. NIR and NIT should be useful as fast screening methods to determine whether beef has been adequately heat treated.     

The use of near infrared reflectance spectroscopy for predicting the quality of grass silage

Near infrared reflectance spectroscopy, using a fixed-filter instrument fitted with 19 filters, was evaluated for the prediction of crude protein and in-vitro dry matter digestibility in commercial farm grass silage. Crude protein (6.5–17.6%) was estimated with acceptable accuracy (standard error of prediction=0.63%) using a six-wavelength calibration, but in-vitro dry matter digestibility (40.7–69.9%) was not predicted with sufficient accuracy (standard error of prediction=2.96%) by this technique.     

Determination of protein concentration in raw milk by mid-infrared fourier transform infrared/attenuated total reflectance spectroscopy

This study investigates the potential use of attenuated total reflectance spectroscopy in the mid-infrared range for determining protein concentration in raw cow milk. The determination of protein concentration is based on the characteristic absorbance of milk proteins, which includes 2 absorbance bands in the 1500 to 1700 cm(-1) range, known as the amide I and amide II bands, and absorbance in the 1060 to 1100 cm(-1) range, which is associated with phosphate groups covalently bound to casein proteins. To minimize the influence of the strong water band (centered around 1640 cm(-1)) that overlaps with the amide I and amide II bands, an optimized automatic procedure for accurate water subtraction was applied. Following water subtraction, the spectra were analyzed by 3 methods, namely simple band integration, partial least squares (PLS) and neural networks. For the neural network models, the spectra were first decomposed by principal component analysis (PCA), and the neural network inputs were the spectra principal components scores. In addition, the concentrations of 2 constituents expected to interact with the protein (i.e., fat and lactose) were also used as inputs. These approaches were tested with 235 spectra of standardized raw milk samples, corresponding to 26 protein concentrations in the 2.47 to 3.90% (weight per volume) range. The simple integration method led to very poor results, whereas PLS resulted in prediction errors of about 0.22% protein. The neural network approach led to prediction errors of 0.20% protein when based on PCA scores only, and 0.08% protein when lactose and fat concentrations were also included in the model. These results indicate the potential usefulness of Fourier transform infrared/attenuated total reflectance spectroscopy for rapid, possibly online, determination of protein concentration in raw milk.     

近红外光谱法鉴别奶牛饲料中三聚氰胺甲醛树脂的可行性研究

收集国内常用的、具有代表性的奶牛精补料33个样品,制备99个三聚氰胺甲醛树脂(MF)掺假样品,在全光谱范围内进行近红外透反射光谱扫描,选择合适的前处理方法,采用BP神经网络方法和PLS-LDA方法分别建立判别模型。建立的BP神经网络判别分析模型的预测正确率为100%,建立的PLS-LDA判别分析模型的交互验证最低错误率为0.0778,模型错分率为0.0667,模型预测错误率为0.1429。说明利用近红外透反射光谱建立定性分析模型来检测奶牛饲料中是否掺有MF的研究是可行的。     

近红外光谱技术快速鉴别原料肉掺假的可行性研究

探讨利用近红外光谱技术结合Fisher两类判别法以及多层感知器(multilayer perceptron,MLP)神经网络快速无损鉴别原料肉是否掺假,并建立多种掺假肉的分类识别模型的可行性.首先近红外结合主成分与Fisher两类判别,建立原料肉与掺假肉的判别函数,以原料肉与注水肉两类样木的平均重心即两类样木的加权平均...     

Simultaneous prediction of acetic acid/ethanol concentrations in their binary mixtures using metalloporphyrin based opto-electronic nose for meat safety applications

This paper presents a comprehensive study on the simultaneous prediction of volatile organic compound (VOC) concentrations in their binary mixtures (acetic acid and ethanol) using partial least square regression (PLSR) and multilayer perceptron neural network (MLP-NN). A metalloporphyrin based opto-electronic nose was developed to record the reflectance from metalloporphyrin sensing film. A ruthenium based metalloporphyrin, 2,3,7,8,12,13,17,18-octaethyl-21H, 23H-porphine ruthenium(II) carbonyl (RuOEPCO), was used as sensing material. The percent change in the reflectance (%ΔR) before and after exposure to different combinations of analyte concentrations were used as the input to the prediction models. The relative standard error of prediction (RSEP, %) of the PLS model was found to be 18.51 and 21.77% for acetic acid and ethanol prediction validated using independent test set, respectively. On the other hand, neural network (multilayer perceptron) produced an average RSEP of 7.27 and 9.13% for acetic acid and ethanol prediction validated using independent test set, respectively. Neural networks produced comparatively lower prediction errors using independent test set validation method and shows potential for further investigation and validation on larger dataset.     

The determination of protein,fat and moisture in bread by near infrared reflectance spectroscopy

Near infrared (n.i.r.) reflectance spectroscopy has been employed for the determination of protein, fat and moisture in sliced white bread. N.i.r. reflectance at six wavelengths was measured using circular samples from each of six alternate slices taken from one half of each of 30 loaves of different composition. The six readings for each loaf at each wavelength were averaged and used to produce calibrations which, on prediction of the compositions of a further 30 loaves sampled in the same way, gave rise to standard deviations of differences between n.i.r. and standard procedures of 0.20% for protein, 0.18% for fat and 0.51% for moisture. Calibrations derived from the other halves of the loaves, which had been air-dried and ground to a powder, resulted in similar standard deviation of differences for protein and fat.