卷烟主流烟气粒相物的电子鼻分析

摘要:建立了主流烟气粒相物的电子鼻分析方法,首先对抽吸方式、粒相物放置时间、延迟时间和进样体积进行优化,然后利用单因素实验对顶空加热温度和顶空加热时间进行优化实验。结果如下:按照口数进行抽吸重现性更好,粒相物的放置时间控制在12h以内,延迟时间为1800s,进样体积为100μL,顶空加热温度和时间分别为90℃和20min。在研究主流烟气粒相物的电子鼻分析方法的基础上,建立了不同风格卷烟的判别模型及同一规格卷烟的质量控制模型,并对模型的可靠性进行了验证。结果发现,所建立的卷烟风格判别模型有效,且判别精度达到了95%;所建立的质量控制模型可以对卷烟质量进行月度跟踪检验,同时该模型在卷烟日常维护方面也有很高的实用价值。      

基于声学特性的鸭蛋破损判别模型研究

本文研制了基于声学特性的鸭蛋破损判别模型。阐述了破损判别试验的试验方法,对获得的试验数据进行处理,提取了表征破损的4个参数,建立Fisher破损判别模型,并对模型进行了显著性检验。试验所得模型可直接用于鸭蛋破损判别。     

基于生成/判别混合模型的动作识别

生成模型能够处理丢失的数据,而判别模型一般具有较好的分类性能。本文提出了一种新的生成/判别混合模型来进行动作识别。该方案利用Fisher核的方法,通过主题模型LDA建立训练样本的Fisher核表示,然后利用核函数训练判别模型进行动作分类。     

基于近红外光谱的卷烟质量投影模型

为提高卷烟配方的稳定性,科学量化卷烟的质量水平,建立基于卷烟产品近红外光谱信息的质量投影模型。通过主成分分析对光谱数据进行降维,利用主成分建立Fisher判别函数对各卷烟产品进行投影。实验结果表明,模型得到的二维投影图能直观反映各卷烟牌号质量的高低关系,为卷烟质量的评价提供了客观的科学依据。     

基于电子舌技术的卷烟主流烟气味觉识别

为考察电子舌对不同卷烟烟气味觉的识别效果,利用电子舌系统检测了6个烤烟型和3个混合型卷烟样品主流烟气水处理液的味觉特征,并对其传感器响应信号进行了主成分分析( PCA)和判别因子分析( DFA).结果表明:①前2维主成分对卷烟品种的味觉识别贡献率达84.82％;②前2维DFA对卷烟品种的味觉识别贡献率达95.42％.结...     

基于遗传算法的卷烟主流烟气粒相物pH值影响因素研究

为研究影响烟气pH值的关键性烟丝化学成分,以国内代表性卷烟为研究对象,分别对样品的烟丝生物碱、糖、氨基酸、有机酸等化学成分含量及主流烟气粒相物pH值进行了测定。采用遗传算法结合偏最小二乘法,筛选出8个对烟气总粒相物pH值影响最大的化学指标,建立了烟气pH值与烟丝化学成分之间的定量关系模型。结果表明,烟丝丁酸、异戊酸、乙酰丙酸、苹果酸、葡萄糖和精氨酸可以降低卷烟烟气pH值,烟碱和异亮氨酸可以增加卷烟烟气pH值。本研究可为卷烟产品研发提供科学依据。      

烟丝裂解产物指纹图谱及在卷烟品质评价中的应用

为考察卷烟烟丝裂解产物应用于卷烟品质评价的可能性,建立了获取卷烟烟丝指纹图谱的裂解-气相色谱/质谱(Py-GC/MS)方法,并对4个品牌共20个卷烟样品的烟丝进行了测定;基于烟丝的Py-GC/MS指纹图谱筛选共有峰,采用聚类分析和主成分分析法对数据进行分析评价,建立了卷烟评价模型,并进行了模型验证。结果表明:①利用Py-GC/MS可对卷烟烟丝进行表征,从烟丝的指纹图谱中共筛选出29个共有峰。②通过模式识别,实现了4个品牌的卷烟样品的较好区分。③建立的评价模型对15个建模样品和5个验证样品的正确判别率为100%。该方法简便、快速,可为卷烟品质评价提供参考。     

电子鼻技术结合化学计量学用于卷烟包装纸VOCs残留量的表征

为建立更加简便快速、环保和低成本的卷烟包装纸品质判别方法,结合卷烟包装纸挥发性有机化合物(VOCs)的质量分数和电子鼻检测结果,用判别因子分析(DFA)、主成分分析(PCA)和簇类独立软模型分析(SIMCA)3种方法分别构建苯残留量判别模型和15种VOCs总量判别模型,并对比了3种方法的建模结果。结果表明:采用DFA构建苯残留量判别模型,以及采用DFA或SIMCA构建15种VOCs总量判别模型,均可较好地实现卷烟包装纸品质的快速判别。     

基于FT-NIR和电子鼻的苹果水心病无损检测

探讨傅里叶变换近红外光谱技术和电子鼻技术应用于苹果水心病检测的可行性。以277?个“秦冠”水心病苹果和健康苹果为试材,分别采集每个样本在12?000～4?000?cm－1波数范围的近红外光谱和10?个传感器的电子鼻信号,用不同预处理的近红外光谱方法提取主成分建立Fisher判别模型;同时电子鼻结合3?种化学计量学的方法进行建模。结果表明,经一阶导数（9?点平滑）预处理的近红外光谱,提取前20?个主成分建立的Fisher判别模型效果最好,对未知样本的正确判别率达100%;电子鼻分别结合Fisher判别、多层感知器神经网络和径向基函数神经网络判别模型对未知样本的识别率为89.7%、89.5%和85.7%。故利用近红外光谱和电子鼻技术分别结合化学计量学的方法可快速、无损检测苹果的水心病。其中,近红外光谱技术结合Fisher判别对苹果水心病的识别率最高,是一种准确可靠的测定方法。     

基于Fisher算法和BP-ANN的造纸原料近红外判别

为了实现对造纸原料的快速准确判别,收集了5种共40份原料品种的近红外光谱数据。通过MAF和一阶导数方法进行光谱数据预处理,用主成分方法对光谱数据进行压缩降维,分别利用Fisher算法和BP人工神经网络来建立原料近红外光谱判别模型,并对两种判别模型进行比较。结果表明:两种模型都能较好地进行造纸原料的近红外判别,且BP人工神经网络比Fisher判别函数在容错性上表现得更为优越,建立的模型用于种类判别时表现得更为稳健。     

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。     

Characterization and Discrimination of the Floral Origin of Sidr Honey by Physicochemical Data Combined with Multivariate Analysis

Sidr honey represents one of the most expensive monofloral honeys worldwide. The quality control of such honey types usually depends on pollen analysis or comparison of physicochemical characters. In the presented work, 38 different honey samples of which 13 represented genuine Sidr (Ziziphus spina-christy) honey samples were collected from various areas of Yemen. All samples were characterized by physicochemical parameters including moisture content, pH, electrical conductivity, and free acidity. The physicochemical data was subjected to multivariate data analysis including principal component analysis (PCA) and hierarchical cluster analysis (HCA). The development of partial least square discriminant analysis (PLS-DA) model on validation gave 100 % correct classification of the test set samples. All tested honey samples were within the level permitted by the international standards for honey quality. The application of the discriminant technique PLS-DA presented excellent potential for discriminating the botanical origin of Yemeni Sidr honey from other non-Sidr samples and may serve as a discriminant model to be applied to other honey types worldwide.     

近红外光谱技术对猪肉注水、注胶的快速检测

采用近红外光谱（near infrared spectroscopy,NIR）结合主成分分析（principal component analysis,PCA） 和判别分析法建立了注水肉、注胶肉和正常肉的定性判别模型。注水肉中注水量的多少对判别准确率产生影响, 当注水量为1.25%～20%时,3 种肉的总体判别准确率为94.23%;当注水量为3.75%～20%时,判别准确率提高至 96.96%。模型对所有预测集样品的总体判别准确率为94.92%。表明NIR结合PCA法、判别分析法判别注水肉、注胶 肉和正常肉具有可行性。采用偏最小二乘法（partial least squares,PLS）结合PCA分别建立了注水量和注胶量的定 量分析模型,经验证,两种模型对预测集样品的预测均方差分别为4.01%和3.87%,预测值与实测值间的相关系数 （r）分别为0.904 2和0.912 8。表明两种模型的预测性能良好。     

不同品种红茶及茶膏的Fisher判别分析

为探究Fisher判别分析法对不同品种红茶及茶膏种类品质区分的可行性,针对不同品种红茶及滇红碎茶茶膏,以蛋白质、总糖、茶红素、茶黄素、茶褐素及9 种矿物质元素为变量进行判别分析。结果表明：采用Fisher判别分析法评定不同品种红茶及不同加工工艺滇红碎茶茶膏的真实归属及品质是可靠的。红茶茶叶及茶膏品质评定中采用Fisher判别分析有利于降低传统感官评定过程中人为主观因素的干扰,避免模糊性和不确定性,保证评审的稳定和可靠性。     

基于气象因子的烤烟香型分区模型构建及应用

为了明确烤烟香型与气候间的关系,以33个不同香型烟叶典型产区的气象数据和烟叶香型为基础,采用逐步判别分析方法,建立了烤烟香型的气象因子评价模型,并利用全国2012年生产的480个烟叶样品对模型进行了验证。结果显示,该模型的判别正确率为86.2%。利用该模型和全国主要烤烟产区的气象因子对各产区烟叶香型的判别结果形成了全国烤烟香型分区图,明确了基于气象因子评价模型的浓香型、中间香型和清香型烟叶的分布区域。      

基于氮稳定同位素比例质谱和液相色谱-高分辨质谱的有机番茄鉴别

为了解决有机食品鉴别需要多种参数指标进行综合判定,制样过程繁琐,判别正确率偏低的问题,本研究提出一种基于稳定同位素比例质谱和液相色谱-高分辨质谱分析技术,结合偏最小二乘法判别分析（partial least squares-discriminant analysis,PLS-DA）的有机番茄鉴别方法。将采集的94 个番茄样品（有机番茄50 个,普通番茄44 个）匀浆破碎、冷冻干燥后,分析并采集氮同位素比值（δ15N）和液相色谱-高分辨质谱数据,应用PLS-DA方法提取数据信息,建立有机番茄判别模型,筛选并确证出其中判别的标志因子,包括氮同位素比值以及有机酸、黄酮类等13 个特征化合物。所建立的模型对有机番茄的判别正确率为93.4%,基本可以满足有机番茄判别的需要。该研究为有机西红柿的快速判定提供了一种新的技术手段。     

Identifying Potato Varieties Using Machine Vision and Artificial Neural Networks

The objective of this study is to develop a method for identifying and discriminating 10 potato varieties by combining machine vision and artificial neural network methods. The potato varieties include Agria, Savalan, Florida, Fontaneh, Natasha, Verona, Karso, Elody, Satina, and Emrad. A total number of 72 characteristic parameters specifying color, textural, and morphological features are found among these varieties. By using principal component analysis, 16 principal features are selected for identifying and discriminating potato varieties. The data obtained from image processing were classified using linear discriminant analysis and non-linear artificial neural network method. The accuracy of discriminant analysis were 73.3, 93.3, 73.3, 40, 73.3, 73.3, 66.7, 80, 40, and 53.3%, respectively, for the varieties used in this study. The classification accuracy was improved by 100% for all the varieties using neural network analysis and the correct classification ratio was 100% using this method. It is revealed from the results that machine vision technique and neural network analysis could identify potato varieties with acceptable accuracy.     

Discrimination and geographical origin prediction of washed specialty Bourbon coffee from different coffee growing areas in Rwanda by using electronic nose and electronic tongue

The study aimed at discriminating washed specialty Bourbon coffee from major coffee growing areas in Rwanda and evaluating the feasibility of using flavor to predict the geographical origin of Arabica coffees from different origins. Discrimination was achieved by performing a principal component analysis, and a discriminant factorial analysis (DFA) model was used to predict the geographical origin of coffee samples based on their intrinsic flavor. Discrimination results from both e-nose and e-tongue indicated clear grouping of coffee samples from areas within the same geographical sub-regions. A DFA model using e-nose was successful in predicting the geographical origin of coffee samples but not with e-tongue. Therefore, the study demonstrated that aroma could reliably be used to predict the geographical origin of coffee samples from different origins than their taste profile.     

Influence of vision systems,black and white,colored and visual digitalization,in natural cork stopper quality estimation

Quality classification of wine natural cork stoppers is related to presence of discontinuities in the cork tissue. Automated image analysis of stoppers based on black and white cameras is used industrially for commercial classification but recently color has been introduced in image processing. This paper compares the performance of three image vision systems regarding classification accuracy of cork stoppers of good, medium and inferior quality: black and white, three‐band RGB color and manual detection by digitalization in color image. A canonical discriminant analysis approach was used to compare the discriminating power between cork stopper quality in each vision system. Good discriminant results were obtained with the area of pores expressed either in total or as ratio, mean or maximum value. The use of color slightly enlarges the range of cork inspection systems and automated systems have a similar accuracy of classification to visual inspection. Copyright © 2007 Society of Chemical Industry