地理标志大米的仿生电子鼻分类识别

为了对地理标志大米进行原产地保护,采用PEN3电子鼻,分析样品质量、项空空间及静置时间等试验参数对电子鼻传感器响应值影响,结合主成分分析(PCA)和线性判别分析(LDA)方法对3个不同地理标志大米进行识别研究。结果表明:选取50 g样品,以50 mL顶空空间、静置1 h测得的电子鼻响应值最佳;PCA法可以区分不同地域的大米,也可以区分不同品种大米,LDA法也可以区分不同地域的大米,但不能区分不同品种的大米。运用电子鼻可以将地理标志大米进行较好的区分,为电子鼻技术应用于大米产地溯源提供理论基础。     

主成分分析法在浓香型白酒酒质评价中的应用研究

为了寻找一种更加客观评价级别相近白酒酒质的方法,本研究利用主成分分析法对相近级别的浓香型白酒色谱成分进行评价,结果发现:主成分分析法能够很好的将级别相近的白酒进行排名,其实际应用有待于深入研究。     

Rapid freshness analysis of mantis shrimps (<Emphasis Type="Italic">Oratosquilla oratoria</Emphasis>) by using electronic nose

In this paper, rapid freshness analysis method of mantis shrimps (MSs) (Oratosquilla oratoria) by using electronic nose (e-nose) was investigated. Shrimps were stored at 4° and ?20°. E-nose responses to samples were measured. Meanwhile, appearance of MS was recorded. Total volatile basic nitrogen (TVB-N) index was examined to provide a standard freshness indicator for samples according to Chinese national standard protocol. E-nose measurement data was processed by principal component analysis (PCA) and non-linear bistable stochastic resonance (SR). Experimental results demonstrated that e-nose sensitively responded to shrimps. PCA results failed to discriminate all shrimps. SR signal-to-noise ratio (SNR) spectrum successfully discriminated all shrimp samples stored at 4° and ?20°. E-nose MS freshness degree forecasting models were developed using SNR maximums. Combining TVB-N examination results, the forecasting accuracy of developed freshness analysis model is 91.67 %. The proposed method had some advantages including rapid analysis, easy operating, non-destructive, low cost, etc.     

Determination of the Freshness of Beef Strip Loins (M. longissimus lumborum) Using Electronic Nose

A beef strip loins (Musculus longissimus lumborum) freshness determination method utilizing electronic nose (e-nose) was investigated in this paper. Fresh beef strip loins samples were stored at 4°C continuously for 10 days. Total viable count (TVC) index, total volatile basic nitrogen (TVB-N) index, and e-nose responses to beef strip loins samples were measured every day. TVC and TVB-N index rose with the increase of storage time. Principal component analysis (PCA) only partially discriminated beef samples under different storage days. Stochastic resonance (SR) signal-to-noise ratio (SNR) spectrum discriminated all beef samples successfully. Beef strip loins freshness discrimination model was developed using SR SNR maximums (SNRmax) linear fitting regression. The proposed method forecasted beef freshness with high accuracy. It is holds promise in meat freshness determination applications.     

A Gaussian-based kernel Fisher discriminant analysis for electronic nose data and applications in spirit and vinegar classification

Fisher discriminant analysis (FDA) is a very useful pattern recognition technique widely used in electronic nose system (e-nose). However, due to its linear characteristic, the classification problems of multi-class and high-dimensional e-nose data cannot be handled effectively. Therefore, a Gaussian-based kernel FDA (KFDA) method is proposed to solve multi-class and high-dimensional classification problems of complex samples such as food classification using e-nose. The key point of the method is how to determine the Gaussian kernel parameter. Firstly, according to distance discriminant analysis viewpoint, a desired kernel matrix adapted to Gaussian kernel function can be given successfully. Secondly, an evaluation function based on Euclidean distance is established for measuring the degree of approximation between actual kernel matrix containing an unknown Gaussian kernel parameter and the desired kernel matrix so as to get an optimal solution of the parameter, and then the actual kernel matrix can be definitely determined. Finally, the principal component analysis (PCA) for the actual kernel matrix is carried out. Meanwhile, FDA for the principal component matrix generated by PCA is also implemented in succession, and the KFDA is completed. Six kinds of Chinese spirit and six kinds of Chinese vinegar samples as two classification applications were respectively carried out accurately with the KFDA method; and the KFDA method is tested to be very simple and effective. The KFDA method may be promising for complex samples classification dataset of e-nose.     

Colorimetric artificial nose for identification of Chinese liquor with different geographic origins

Colorimetric artificial nose was used to characterize and identify Chinese liquors from six different geographic origins. Using chemical dyes as the sensing elements, the developed colorimetric artificial nose showed a unique pattern of color changes upon its exposure to Chinese liquors. Data analysis was performed by chemometric techniques: Hierarchical cluster analysis (HCA), principal component analysis (PCA) and linear discriminant analysis (LDA). Each category of Chinese liquor could cluster together in PCA score plot. No errors in classification by HCA were observed in 45 trials. LDA model showed a 100% of prediction ability for Chinese liquor. The results demonstrated that colorimetric artificial nose was able to classify Chinese liquors from different geographic origins.     

基于电子鼻/舌融合技术的白酒类别辨识

为实现白酒质量的快速鉴别,本研究采用两个不同的电子鼻和电子舌融合系统采集不同品牌白酒样品的气-味信息,采用主成分分析法、K均值法对检测结果进行聚类分析,采用支持向量机法对白酒的品牌进行预测分类分析。经主成分分析聚类分析后,应用基于TGS型气敏传感器的电子鼻和电子舌融合系统可以将3种酒进行很好的区分,其余5种酒有交叉,而应用基于MQ、MP型气敏传感器的电子鼻和电子舌融合系统可以将8种白酒基本区分开来;K均值法分别用于两融合系统,前融合系统的错分类概率为33.3%,后融合系统的错分类概率为23.75%。采用支持向量机法对白酒的品牌进行预测分类,应用前融合系统的识别率为93.75%,而应用后融合系统的识别率为98.75%。结果表明:气-味信息融合技术可以实现对白酒品牌的鉴别,且应用基于MQ、MP型气敏传感器的电子鼻和电子舌融合系统对白酒类别的识别结果较好于应用基于TGS型气敏传感器的电子鼻和电子舌融合系统。     

Discrimination of different kinds of Luzhou-flavor raw liquors based on their volatile features

The volatile constitutions of 36 raw liquors from three distilling stages (head, heart and tail) of two typical Luzhou-flavor liquors (Fenggu-FG and Jiannanchun-JNC) were identified and semi quantified by gas chromatography-mass spectrometry (GC-MS). A total of 63 compounds were identified in all liquors. These two typical liquors had similar volatile constitutions, in which, 3-methylbutanol, hexanoic acid, ethyl hexanoate, ethyl butanoate, ethyl lactate, ethyl pentanoate, 1,1-dimethoxythane and 1,1-diethoxy-3-methylbutane were considered to be main compounds due to their high concentrations. Multivariate analyses including principal component analysis (PCA) and partial least square-discrimination analysis (PLS-DA) were conducted to reveal the detailed distinctions of liquors with different origins and reveal the volatile markers of several kinds of liquors. PCA explanation plane primary revealed the main differentiation between FG and JNC based on their loading plot values on axis PC1. Results of PLS-DA showed the detailed distinctions of liquors, suggesting that 2-methylpropanoic acid, butanoic acid, pentanoic acid, nonanoic acid, 2,3-butanediol, 1-hexanol, and ethyl nonanoate strongly correlated with FG liquors, while n-butylformate, isopentyl butanoate, isoamyl caproate, and p-cresol contributed to the specificity of JNC liquors. Furthermore, differences amongst the heart distilling stage liquors from different enterprises were also visualized in the two-dimensional PLS-DA discrimination plane. To our knowledge, this is the first article using the GC-MS paired with multivariate analysis to discriminate different kinds of Luzhou-flavor raw liquors.     

荧光素对中国典型香型白酒的荧光鉴别

采用荧光光谱技术与模式识别分析法建立一种快速区分白酒香型的新技术。在495nm可见光激发下分别测定6种香型白酒、荧光素及这两种物质作用后200～800nm之间的荧光光谱．探讨了荧光素与白酒的作用机理。结合模式识别法对310～380nm和504～530nm两个波段对应的荧光强度值进行分层聚类分析（Hierarchieal Cluster Analysis．HCA）和线性判别分析（LinearDiscriminantAnalysis,LDA）,对荧光素区分不同香型白酒的效果进行验证。结果表明,HCA分析能完全区分6种香型白酒,LDA分析对所有样本均得到正确的判别．正确率为100％。因此,这种基于荧光素的荧光光谱法简单、有效,可用于不同香型白酒的快速区分。     

Application of Vis/NIR Spectroscopy for Chinese Liquor Discrimination

As one of the most widely consumed alcoholic beverages, Chinese liquor varies greatly in price, flavor, and quality. This diversity calls for effective and reliable discrimination methods. In an attempt to find the best liquor discrimination method, this study used different methods to analyze and identify 730 Chinese liquor samples including 22 kinds, ten brands, and six flavors. These samples, covering most of the famous liquors in China, were analyzed by visible and near-infrared (Vis/NIR) spectroscopy and modeled by three classification methods including supporting vector machine, soft independent modeling of class analogy, and linear discriminate analysis based on principal component analysis (PCA-LDA). Pretreatments and parameters for each model were optimized, and models discrimination ability was compared. The research finds that PCA-LDA was the best model with an average prediction rate of 98.94 % in the training set and 95.70 % in the test set. The correct rates for brands, flavor styles, ages, and alcohol degrees were all higher than 95 %. It shows that Vis/NIR is a reliable, inexpensive, and effective tool for Chinese liquors discrimination.     

Electronic nose system fabrication and application in large yellow croaker (<Emphasis Type="Italic">Pseudosciaena crocea</Emphasis>) fressness prediction

In this paper, an electronic nose (E-nose) system was fabricated, and its application in large yellow croaker (Pseudosciaena crocea) freshness prediction was also explored. E-nose responses to samples stored at 277 K were measured for 8 days. Freshness indexes, such as total viable counts (TVC), total volatile basic nitrogen (TVB-N) and K value, were synchronously examined by chemical examinations. Principal component analysis (PCA) and stochastic resonance (SR) were utilized for e-nose data analysis. Results suggested that PCA showed poor freshness discrimination result. SR signal-to-noise (SNR) spectrum using maximal SNR (\(Max_{SNR}\)) values quantitatively characterized freshness of all croakers. Multiple variable regression (MVR) result demonstrated that there was good linearity relationship between SR \(Max_{SNR}\) values and fish freshness indexes. Large yellow croaker freshness predicting model was developed by non-linear fitting regression on \(Max_{SNR}\) values with high accuracy and repeatability. Therefore, the method proposed in this paper provides a rapid and nondestructive methodology for freshness prediction of large yellow croakers. The predicting error of the developed model is 10 %.     

Assessment of Shelf Lives of Black Pepper and Small Cardamom Cookies by Metal Oxide-Based Electronic Nose Using Spoilage Index

E-nose technology has been successfully employed in determination of rancidity in nutraceutical-rich drop cookies. The cookies were formulated with extracts obtained by supercritical carbon dioxide extractions and post-extraction sample matrices of black pepper and small cardamom. Rancidity in cookies was estimated by comparing the odor profiles of stored cookies with the profiles of deliberately rancid cookies (training sets), using e-nose. PCA plots along with spoilage indices obtained from Mahalanobis distances calculated from e-nose responses revealed that the extracts and post-extraction sample matrices of these spices performed as natural antioxidants in preventing rancidity in cookies. Both the extracts enhanced the shelf lives of cookies by at least 120 days; post-extraction matrices of black pepper and small cardamom enhanced the same by 80 and 40 days, respectively. These findings were further affirmed by biochemical analyses and linear regression equations were generated for prediction of FFA content, PV, and MDA values from the spoilage indices of each type of cookie. Thus, spoilage indices can circumvent the requirement of conducting cumbersome biochemical assays for estimation of rancidity in cookies. Similar applications of e-nose technology and spoilage indices can be envisaged for other bakery products also.     

Monitoring of evolution during red wine aging in oak barrels and alternative method by means of an electronic panel test

A combination of the data obtained by means of an e-nose (based on resistive MOX sensors), an e-tongue (based on voltammetric sensors) and an e-eye (based on CIE Lab coordinates) has been used to monitor the aging of a red wine. The changes in the chemical composition of wines that occur during maturing have permitted the system to discriminate among wine samples collected after one, three, six and ten months of aging. The discrimination capability of the electronic panel test obtained by means of Principal Component Analysis (PCA), Partial Least Squares Discriminant Analysis (PLS-DA) is even higher than the discrimination achieved by means of traditional chemical analysis.After ten months of aging it has been possible to discriminate between the wine aged in a French oak barrel and the same wine soaked with oak chips of the same origin and toasting level and treated with microoxygenation.     

主成分分析研究白酒基酒香气成分

以485个优质基酒和普通基酒的主要香气成分数据为基础,采用R语言软件,分别对两类基酒进行4次随机抽样,每次抽取60个样品,基酒品质分析数据经标准化处理后进行主成分分析。结果表明,优质基酒4次抽样主成分分析结果比较稳定,普通基酒分析结果差异较大。碎石图检验表明,优质基酒在第一、二、三主成分后断崖明显,普通白酒在第一主成分后下降平缓,没有明显断崖。分布分析表明,优质基酒密度分布图的尖峰厚尾现象明显,方差为0.560,样本间的差异小,香气成分含量范围相对集中;普通基酒分布较为分散,方差为0.925,样本间差异较大,香气成分含量范围不稳定。在白酒风味分析时,应考虑样本的数据特征,以保证分析结果的准确性和可靠性,进而有效指导白酒生产。     

一种可视化快速鉴别白酒的新方法研究

白酒主体香味成分的含量和比例,是白酒香型风格的"构成要素"。针对白酒的主体香味成分,选择筛选出特异性敏感元件20种,构建5×4比色传感阵列芯片,建立了一种可视化快速鉴别白酒新方法。新阵列芯片检测性能明显优于已报道的通用阵列,可精确识别出不同风味白酒,且能反映出不同白酒中主体香味成分的差异。对15种白酒进行检测,主成分分析结果表明,前三个主成分对识别的贡献率依次为酯、醛、缩醛类物质(34.5%),酸类物质(19.2%)和乙醇(11.9%)。聚类分析发现同一品牌的白酒能正确归类,不同品牌同种香型白酒能率先聚为一类。各白酒间的相似性和差异性在聚类图上有较好的体现,不同香型白酒间的距离远近与白酒的生产原料、糖化发酵剂、酿造工艺密切相关。     

基于无机元素含量的地理标志食醋分类

采用ICP-AES和ICP-MS共测定了山西老陈醋、镇江香醋中38个无机元素的含量。经显著性分析(T检验)筛选出Cd、Al、Ca、Ba、Fe、Mn、Zn、Ni、Pb、Cs、As、Mo等12个在山西老陈醋和镇江香(陈)醋中含量差异最大的特征元素。基于以上12个特征元素,采用主成分分析法对样品进行分类,可明显归于两类,其含量可能与产地的质地条件和产品的生产工艺有关。无机元素含量分析,有望成为一种鉴别地理标志保护食醋的技术手段。     

Nondestructive Differentiation of Panax Species Using Visible and Shortwave Near-Infrared Spectroscopy

Visible and short-wave near-infrared (Vis-SWNIR) spectroscopy was investigated to differentiate species of panax, including American Panax quinquefoliumI, Chinese Panax quinquefolium, and Chinese Panax ginseng. Principal component analysis (PCA) was applied before least-square support vector machine (LS-SVM) modeling, and the vast points of the spectral data (376–1,025 nm) were effectively reduced. PCA-LS-SVM differentiated species with 100% correct classification rate for the tested samples. In addition, effective wavelengths were selected according to modeling power, discrimination power, regression coefficients, loading weights, and genetic algorithms, respectively. The optimal and simplified LS-SVM model with 100% correct classification rate was achieved using the effective wavelengths selected by genetic algorithms. The results showed that Vis-SWNIR spectroscopy technique can be applied as a high accuracy and fast way for the qualitative discrimination of herb species.     

基于低场核磁共振的白酒品牌判别分析

为探求一种白酒品牌判别的方法,基于低场核磁共振（LF-NMR）技术,综合运用主成分分析（PCA）、偏最小二乘法-判别分析（PLS-DA）、反向传播人工神经网络（BP-ANN）方法对6种浓香型白酒品牌共300个样本进行模式识别分析,解析了对不同品牌浓香型白酒进行判别的可行性。结果表明,运用PCA方法对样品进行识别,无法区分白酒品牌;运用PLS-DA的方法对白酒样品进行识别,训练集的识别率约为99.5%,预测集识别率约为96.7%;运用BP-ANN的方法对白酒样品进行识别,训练集识别率约为99.5%,预测集识别率约为98.9%。结果表明,PLS-DA方法和BP-ANN方法对浓香型白酒样品的区分成功,表示将低场核磁共振方法应用到浓香型白酒的品牌判别中是可行有效的。     

Analysis of ethanol in soy sauce using electronic nose for halal food certification

The purpose of this study was to analyze ethanol content in soy sauce using mass spectrometry (MS) with electronic nose (e-nose) to determine if MS e-nose can replace gas chromatographic analysis for halal certification. Gas chromatography–flame ionization detector (GC-FID), the standard method of ethanol content, was used to analyze 24 different kinds of soy sauce. Ethanol was detected from 13 soy sauce samples in the concentration range of 0.0004–1.7wt%. The MS e-nose data were analyzed by discriminant function analysis (DFA). Based on an addition method, the results were more than 96.6% accurate when the ethanol concentrations were greater than 0.5%. A high correlation between the first score of the DFA plot and the ethanol concentration was observed. Thus, mass spectrometry based on e-nose is an efficient method for determining ethanol as a primary screening tool for halal certification.