基于拉曼光谱和CNN算法的特级初榨 橄榄油的掺假量化

旨在为特级初榨橄榄油掺假快速定量分析提供参考,以掺假菜籽油的特级初榨橄榄油为例,采用激光拉曼光谱实验系统获取油样的拉曼光谱数据,运用基于Inception V2结构的卷积神经网络（CNN）算法提取拉曼光谱特征并完成光谱特征与掺假量的非线性关系映射。结果表明：特级初榨橄榄油与菜籽油的拉曼光谱存在较大的差异,其中类胡萝卜素、碳碳双键、甲基和亚甲基产生的拉曼特征峰是引起差异的主要因素;所建立的CNN模型效果较好,训练集、验证集、测试集的决定系数均大于099,均方根误差均小于0.026;在低剂量掺假中,模型的预测结果仍具有一定的参考价值。综上,拉曼光谱结合基于Inception V2结构的CNN算法所建立的模型可以满足特级初榨橄榄油掺假量的快速检测。     

基于光谱法的特级初榨橄榄油快速鉴伪技术

本文研究了特级初榨橄榄油中掺入不同比例橄榄果榨油(精炼橄榄油)、菜籽油、玉米油和大豆油的光谱特征,采用荧光光谱和紫外光谱,对掺假样品及纯油样品进行了快速检测。结果表明,特级初榨橄榄油的光谱特征与其他植物油之间差异较大,且掺假体积与吸光度之间存在良好的线性关系(R²>0.89),实现了特级初榨橄榄油的定性鉴别与定量检测,建立了特级初榨橄榄油质量控制体系及其掺假检测分析技术,最低检出限为1%,线性范围为5%～100%(v/v)。系统聚类分析将所有特级初榨橄榄油准确地分为一个亚类,也佐证了此方法的稳定性与可靠性。这种简单快捷的检测技术,有助于特级初榨橄榄油实时、在线橄榄油检测分析技术的研发,为我国橄榄油品质鉴定及产业发展提供有利的技术保障。     

油脂同步荧光光谱检测及橄榄油掺假鉴别

采用同步荧光光谱仪,在激发波长250~720 nm,波长间隔Δλ=15 nm时,采集20种食用植物油和掺杂的特级初榨橄榄油的荧光光谱图,分析比较了各种植物油脂的同步荧光光谱图。结果表明,同步荧光光谱法能够将特级初榨橄榄油与其他17种植物油明显地区分开来。在橄榄油掺杂鉴别中,其中14种植物油掺兑量在1%的情况下,同步荧光光谱图与特级初榨橄榄油有着明显的差异。同步荧光光谱法对橄榄油掺假鉴别,无需复杂的样品前处理,本方法简便、快速、灵敏,适合快速筛查。     

基于PCA-SVM结合共聚焦拉曼光谱的特级 初榨橄榄油掺伪压榨菜籽油定量分析

为了促进国内橄榄油市场的健康发展，对掺伪同样存在天然类胡萝卜素的低温压榨菜籽油的特级初榨橄榄油进行了定量鉴别研究。采用共聚焦拉曼光谱技术对不同掺伪浓度油样进行测试，基于密度泛函理论对油样的拉曼光谱峰的归属进行了理论分析，并对拉曼光谱数据进行主成分分析（PCA），然后利用支持向量机（SVM）构建PCA-SVM模型。另外，对PCA-SVM模型的检出限进行了研究。结果表明：特级初榨橄榄油与低温压榨菜籽油的拉曼光谱存在一定差异，最明显的光谱差异主要集中在谱峰1 008、1 161、1 528 cm-1和谱段2 800～3 000 cm-1内，与密度泛函理论对不同油样拉曼光谱峰的分析一致；不考虑类胡萝卜素特征信号建立的PCA-SVM模型决定系数大于0.989，均方根误差小于2.990%，检出限为2%（低温压榨菜籽油体积分数）;在特级初榨橄榄油掺伪定量分析中，考虑类胡萝卜素的特征信号有助于提高模型预测精度，但仅限于掺伪低价植物油中无类胡萝卜素存在的情况;PCA-SVM模型在不考虑类胡萝卜素特征信号的情况下依然具有良好的定量预测效果。综上，所建立的PCA-SVM模型可以用于掺伪2%以上低温压榨菜籽油的特级初榨橄榄油的定量鉴别。     

基于多元素含量分析结合化学计量学技术的橄榄油等级鉴别方法

基于多元素含量分析结合化学计量学技术对特级初榨橄榄油和精炼橄榄油进行等级鉴别。结果表明,采用主成分分析可成功区分特级初榨橄榄油和精炼橄榄油,建立可靠的橄榄油等级鉴别的正交偏最小二乘法判别分析模型,结果与主成分分析一致,且各等级橄榄油组内聚类效果明显;聚类分析也有效鉴别了两种等级橄榄油。筛选出区分两个等级橄榄油的8 种特征性元素,分别为Al、Fe、Cu、Ba、V、Sc、La、Zn。因此,采用多元素含量分析结合化学计量学技术可用于特级初榨橄榄油和精炼橄榄油的鉴别。     

三维荧光光谱法快速检测橄榄油中掺假廉价油

目的 建立三维荧光光谱结合机器学习快速检测橄榄油中掺假廉价油的方法。方法 采集橄榄油及掺入大豆油、玉米油、棕榈油三种不同浓度梯度油的荧光光谱数据,利用标准差标准化（standardscaler）、标准正态变换（standard normal variate,SNV）、归一化（normalize）三种光谱预处理方法,基于K近邻(K-nearest neighbor,KNN)、随机森林(random forest,RF)、支持向量机(support vector machine,SVM)、偏最小二乘法(partial least squares,PLS)和卷积神经网络(convolutional neural network,CNN) 5种机器学习方法,构建5种橄榄油定量掺假模型。结果 在定性模型中,基于PLS算法构建的模型效果最好,对3种掺假橄榄油的准确率为79%~97%,其中,在鉴定掺假大豆油的橄榄油中正确率高达97%。在构建的掺假油定量模型中,Standardscaler预处理结合RF算法,构建的定量模型最优,Rc2、Rp2、RMSEC、RMSEP最高,分别为1.00、0.99、0.01、0.02。结论 构建橄榄油掺假3种油的定性定量模型,并建立一种快速、实时、低成本的橄榄油掺假检测方法,能够准确判断是否掺入廉价油,并量化掺假程度,提供更全面的橄榄油质量评估。     

拉曼光谱结合光谱特征区间筛选算法快速定量鉴别植物调和油品质

本研究提出了一种基于拉曼光谱与光谱特征区间筛选算法实现植物调和油中高价值植物油含量快速定量检测的方法。首先,将粒子群优化（particle swarm optimization,PSO）算法与灰狼优化（grey wolf optimization,GWO）算法融合构建混合智能优化算法,即PSOGWO算法。其次,将PSOGWO与组合移动窗口（combined moving window,CMW）策略结合构建新型的拉曼光谱特征区间筛选算法,即PSOGWO-CMW算法。然后,将玉米油（corn oil,CO）和特级初榨橄榄油（extra virgin olive oil,EVOO）以不同比例配制为CO-EVOO植物调和油,并采集其拉曼光谱。将拉曼光谱输入偏最小二乘回归、PSO-CMW、GWO-CMW和PSOGWO-CMW模型预测EVOO含量,并比较建模效果。结果表明,PSOGWO-CMW模型具有最佳的预测性能。采用本方法与气相色谱-质谱法分别检测真实的CO-EVOO植物调和油样本中EVOO含量,结果表明两者的检测性能无显著差异。本方法快速、准确,亦可用于其他植物调和油中高价值植物油含量的快速...     

基于荧光光谱的初榨椰子油掺假检测技术研究

分析初榨椰子油、葵花籽油、玉米油和大豆油的荧光特性,建立SIMCA和PLS-DA判别分析模型,进行初榨椰子油的掺假检测分析研究。结果发现,初榨椰子油与葵花籽油等高不饱和型植物油的荧光特性具有显著性差异,且掺假初榨椰子油的荧光强度与掺假质量分数具有较强的相关性。采集350~400 nm的荧光光谱建立判别分析模型,结果显示,PLS-DA模型具有较好的判别分析能力,其对初榨椰子油中葵花籽油、玉米油和大豆油的掺假识别率分别为96%,90%和93%,该方法可以进行初榨椰子油掺假检测,具有无需前处理、快速、简便、易操作等特点。     

便携式激光拉曼光谱法快速鉴别橄榄油掺假

目的采用便携式激光拉曼光谱仪,建立激光拉曼光谱对橄榄油进行快速鉴别的方法。方法对橄榄油样品进行光谱扫描及基线校正后,以1440 cm-1作为参考波数,对拉曼光谱数据进行归一化处理。结果对80余份橄榄油样品进行统计分析,发现75%的样品在1265 cm-1的拉曼光谱强度值低于540。特级初榨橄榄油中掺加果渣油,会使1265 cm-1和1650 cm-1的特征峰增强,1525 cm-1处的精细结构变小直至消失。结论拉曼光谱具有便捷、快速、无损分析的特点,可作为橄榄油真伪鉴别在线初步筛查的工具。     

豆甾二烯用于特级初榨橄榄油掺假检测的研究

研究了利用豆甾二烯进行特级初榨橄榄油的掺假检测,考察了在特级初榨橄榄油中掺入不同比例油橄榄果渣油、榛子油、葵花籽油、大豆油、玉米油、花生油、米糠油、棕榈油和核桃油,采用高效液相色谱法(HPLC)测定掺和油中的3,5-豆甾二烯含量。植物油样品用石油醚溶解,过硅胶柱净化,C30柱分离,乙腈/叔丁基甲醚(75∶25,V/V)作流动相,紫外检测波长235nm。结果表明,本方法具有良好的回收率及精密度,3,5-豆甾二烯是特级初榨橄榄油掺假鉴别的一个十分重要的特征性指标,本方法能够鉴别出了特级初榨橄榄油中掺入0.3%油橄榄果渣油、0.5%榛子油、2.1%葵花籽油、0.5%大豆油、0.3%玉米油、1.5%花生油、0.2%米糠油、0.6%棕榈油和0.6%核桃油。     

基于三维荧光光谱的花生油掺伪快速检测研究

建立一种基于三维荧光光谱的花生油掺伪检测方法。以纯花生油和掺伪4种常见植物油的花生油为研究对象,将三维荧光光谱图处理转化为灰度图,利用Zernike图像矩直接提取三维荧光光谱灰度图的特征信息,得到的特征信息数据通过Xgboost算法和广义回归神经网络（GRNN）算法分别建立定性和定量掺伪判别模型并对其进行验证。结果表明：Xgboost算法可以有效地对掺伪的花生油进行鉴别,并准确解析其掺伪具体成分;GRNN算法可定量预测花生油掺伪含量,各检出限分别为掺伪大豆油0.2%、掺伪菜籽油1.5%、掺伪玉米油1.0%、掺伪葵花籽油0.5%。因此,该方法可对花生油掺伪进行定性和定量分析,具有快速、简便、灵敏度高等优点。     

紫外光谱结合化学计量学检测初榨橄榄油掺伪研究

以紫外光谱为技术手段,结合偏最小二乘法和BP人工神经网络2种化学计量学方法建立了初榨橄榄油/混合橄榄油二元掺伪体系的定量预测模型.试验结果表明,2种统计模型定量预测性能良好,偏最小二乘模型的训练集交叉验证均方根误差RMSEcv和预测集均方根误差RMSEP均达到0.011,预测值与真实值相关性达到0.996 2;BP人工神经网络迭代次数为61步,训练集拟合残差为9.684×10-5,网络预测值和真实值相关系数为0.998 3,对于5％以上掺伪比例的油样BP神经网络能够精确地预测.     

Authentication of Olive Oil Adulterated with Vegetable Oils Using Fourier Transform Infrared Spectroscopy

Determination of the authenticity of extra virgin olive oils has become more important in recent years following some infamous adulteration and contamination scandals. The study focused on application of Fourier transform infrared spectroscopy to identify the adulteration of olive oils. Single-bounce attenuated total reflectance measurements were made on pure olive oil and olive oil samples adulterated with varying concentrations of sunflower oil (20-100 mL vegetable oil/L of olive oil). Discriminant analysis using 12 principal components was able to classify the samples as pure and adulterated olive oils based on their spectra. A partial least squares model was developed and used to verify the concentrations of the adulterant. Furthermore, the discriminant analysis method was used to classify olive oil samples as distinct from other vegetable oils based on their infrared spectra.     

基于同步荧光光谱的杜仲籽油掺假检测研究

建立基于同步荧光光谱的杜仲籽油掺假判别分析模型及检测方法。以杜仲籽油和7种常见植物油为研究对象,采集激发波长范围为250～700 nm,波长间隔为60 nm的同步荧光光谱,分析杜仲籽油和常见食用油的荧光光谱特性,利用光谱峰面积建立掺假判别模型并对其进行验证。结果表明:杜仲籽油与其他7种植物油的荧光特性存在显著差异;分别利用600～700 nm和300～500nm波长范围同步荧光光谱进行主成分分析,其对杜仲籽油掺假识别准确率高达100%;利用峰面积与掺假比例建立定量判别分析模型,检测限分别为1%和0. 48%。该方法可实现对杜仲籽油掺假的定性和定量分析,且具有较高的灵敏度、简便和快速等特点。     

Fast Fluorescence Spectroscopy Methodology to Monitor the Evolution of Extra Virgin Olive Oils Under Illumination

Detailed regions of excitation and emission wavelengths for extra virgin olive oil samples have been extracted from three dimensional front-face fluorescence spectra. Information was applied to establish a right-angle fluorescence procedure. A right-angle device was assembled and tested with simultaneous excitation from 200 to 400 nm and registration of the fluorescence signal emitted from 400 to 850 nm. A principal component analysis was performed on the signal ranging from 400 to 550 nm from spectra of olive oils officially categorized as extra virgin in order to model the expected variability of compounds related to oxidative processes. Such model was useful to monitor the spectral evolution of extra virgin olive oil samples acquired at retail markets, which were exposed to indirect light during 2 months, through the analysis of the effect on their scores. Three relevant peaks characterized such evolution, with local maxima at around 434 to 437, 464 to 469 and 510 to 518 nm. Polynomial relationship was found between the evolution of those peaks and that of the chlorophyll, at around 670 to 673 nm, with R ² values of 0.98 and 0.99.     

Infrared spectroscopy for quantitative analysis and oil parameters of olive oil and virgin coconut oil: A review

Vegetable oils are major lipid sources with high nutritional and calorific values for human diet. Specifically, virgin coconut oil and extra virgin olive oil are the functional oils widely used in food and pharmaceutical products, either as vehicles or main components. The quality of edible oils is determined by its contents and parameters inherent in vegetable oils. Infrared spectroscopy is an ideal technique for quantitative analysis of vegetable oils as well as for determination of oils parameters as the changes in infrared spectra can be associated with the changes of oils parameters. Infrared spectra in complex samples are difficult to interpret, as a consequence, spectroscopist uses additional tools called with chemometrics to analyse edible oils qualitatively and quantitatively. This article reviews the use of infrared spectroscopy combined with chemometrics (multivariate analysis) for quantitative analysis and determination of oil parameters of virgin coconut oil and extra virgin olive oil. Although infrared spectra for edible oils are similar, they exhibit some differences which enable spectroscopist to differentiate due to the nature property of infrared spectroscopy spectra as fingerprint spectra which can be understood that there are no different edible oils having the same infrared spectroscopy spectra.