Use of chemical profiling to differentiate geographic growing origin of raw pistachios

The objective of this study was to demonstrate the feasibility of chemical profiling methods combined with multivariate methods to differentiate the geographical growing regions of pistachios (Pistachia vera). Elemental analysis (Ba, Be, Ca, Cu, Cr, K, Mg, Mn, Na, V, Fe, Co, Ni, Cu, Zn, Sr, Ti, Cd, and P) of pistachios samples was performed using inductively coupled plasma atomic emission spectrometry. Analysis of inorganic anions and organic acids (selenite, bromate, fumarate, malate, selenate, pyruvate, acetate, phosphate, and ascorbate) of pistachio samples was performed using capillary electrophoresis. Bulk carbon and nitrogen isotope ratios were performed using stable isotope MS. There were nearly 400 pistachio samples analyzed from the three major pistachio growing regions: Turkey, Iran, and California (United States). A computational evaluation of the trace element data sets was carried out using statistical pattern recognition methods including principal component analysis, canonical discriminant analysis, discriminant analysis, and neural network modeling. Several linear discriminant function models classified the data sets with 95% or higher accuracy. We report the development of a method combining elemental analysis and classification techniques that may be widely applied to the determination of the geographical origin of foods.     

Chemical profiling with modeling differentiates wild and farm-raised salmon

Classifications of fish production methods, wild or farm-raised salmon, by elemental profiles or C and N stable isotope ratios combined with various modeling approaches were determined. Elemental analysis (As, Ba, Be, Ca, Co, Cd, Cr, Cu, Fe, K, Mg, Mn, Na, Ni, P, Pb, Sr, Ti, and Zn) of wild and farm-raised salmon samples was performed using an inductively coupled plasma atomic emission spectroscopy. Isotopic and compositional analyses of carbon and nitrogen were performed using mass spectrometry as an alternative fingerprinting technique. Each salmon (king salmon, Oncorhynchus tshawytscha ; coho salmon, Oncorhynchus kisutch ; Atlantic salmon, Salmo salar ) was analyzed from two food production practices, wild and farm raised. Principal component analysis (PCA) and canonical discriminant analysis (CDA) were used for data exploration and visualization. Five classification modeling approaches were investigated: linear discriminate function, quadratic discriminant function, neural network, probabilistic neural network, and neural network bagging. Methods for evaluating model reliability included four strategies: resubstitution, cross-validation, and two very different test set scenarios. Generally speaking, the models performed well, with the percentage of samples classified correctly depending on the particular choice of model and evaluation method used.     

Chemical profiling to differentiate geographic growing origins of coffee

The objective of this research was to demonstrate the feasibility of this method to differentiate the geographical growing regions of coffee beans. Elemental analysis (K, Mg, Ca, Na, Al, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Mo, S, Cd, Pb, and P) of coffee bean samples was performed using ICPAES. There were 160 coffee samples analyzed from the three major coffee-growing regions: Indonesia, East Africa, and Central/South America. A computational evaluation of the data sets was carried out using statistical pattern recognition methods including principal component analysis, discriminant function analysis, and neural network modeling. This paper reports the development of a method combining elemental analysis and classification techniques that may be widely applied to the determination of the geographical origin of foods.     

Coupling proton transfer reaction-mass spectrometry with linear discriminant analysis: a case study

Proton transfer reaction-mass spectrometry (PTR-MS) measurements on single intact strawberry fruits were combined with an appropriate data analysis based on compression of spectrometric data followed by class modeling. In a first experiment 8 of 9 different strawberry varieties measured on the third to fourth day after harvest could be successfully distinguished by linear discriminant analysis (LDA) on PTR-MS spectra compressed by discriminant partial least squares (dPLS). In a second experiment two varieties were investigated as to whether different growing conditions (open field, tunnel), location, and/or harvesting time can affect the proposed classification method. Internal cross-validation gives 27 successes of 28 tests for the 9 varieties experiment and 100% for the 2 clones experiment (30 samples). For one clone, present in both experiments, the models developed for one experiment were successfully tested with the homogeneous independent data of the other with success rates of 100% (3 of 3) and 93% (14 of 15), respectively. This is an indication that the proposed combination of PTR-MS with discriminant analysis and class modeling provides a new and valuable tool for product classification in agroindustrial applications.     

Determining the geographic origin of potatoes with trace metal analysis using statistical and neural network classifiers

The objective of this research was to develop a method to confirm the geographical authenticity of Idaho-labeled potatoes as Idaho-grown potatoes. Elemental analysis (K, Mg, Ca, Sr, Ba, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Mo, S, Cd, Pb, and P) of potato samples was performed using ICPAES. Six hundred eight potato samples were collected from known geographic growing sites in the U.S. and Canada. An exhaustive computational evaluation of the 608 x 18 data sets was carried out using statistical (PCA, CDA, discriminant function analysis, and k-nearest neighbors) and neural network techniques. The neural network classification of the samples into two geographic regions (defined as Idaho and non-Idaho) using a bagging technique had the highest percentage of correct classifications, with a nearly 100% degree of accuracy. We report the development of a method combining elemental analysis and neural network classification that may be widely applied to the determination of the geographical origin of unprocessed, fresh commodities.     

Wavelength Selection for Classifying Paddy Rice by Near‐Infrared Spectroscopy

Using five paddy rice cultivars grown in Central, Eastern, and Southern Taiwan and harvested in the summers of 1997, 1998, and 1999, eight calibrated models were established by discriminant analysis and back‐propagation neural network with four wavelength selection methods. Randomly adding 80 samples of the 2000 year crop in the three‐crop‐year calibrated models for annual recalibration, eight models were used to classify paddy rice harvested in the summer of 2000. With 351 wavelengths of models 1 and 2, the average classification rates by discriminant analysis and backpropagation neural network were 98.1 and 92.5%, respectively. With 69 wavelengths selected by stepwise discrimination of models 3 and 4, the average classification rates by discriminant analysis and backpropagation neural network were 98.5 and 85.5%, respectively. With 69 wavelengths selected by correlation matrix of models 5 and 6, the average classification rates by discriminant analysis and neural network were 72.0 and 72.2%, respectively. With 69 wavelengths from loading values in the first and second principal components of models 7 and 8, the average classification rates by discriminant analysis and neural network were 69.1 and 60.6%, respectively. Model 3 would be recommended for classifying paddy rice to set trading prices because of its highest classification rate (98.5%).     

Saskatoon and wild blueberries have higher anthocyanin contents than other Manitoba berries

Studies have shown that anthocyanins present in berry fruits have some beneficial health effects such as reducing age-associated oxidative stress and possessing anti-inflammatory properties. Therefore, six Manitoba berries (wild blueberry, Saskatoon berry, raspberry, chokecherry, strawberry, and seabuckthorn) were studied for their anthocyanin compositions (mg/100 g) on dry weight basis. Saskatoon berry and wild blueberry showed a high content of total anthocyanins (562.4 and 558.3 mg/100 g, respectively) that were not significantly (P>0.05) different from each other. The corresponding values for other berries: raspberry (365.2 mg/100 g), chokecherry (177.39 mg/100 g), and strawberry (97.5 mg/100 g) were significantly different from each other (P<0.05), and the total anthocyanin content of seabuckthorn was negligible (0.84 mg/100 g). Fifteen major anthocyanins were isolated from Manitoba berries. Saskatoon berry and wild blueberry contained higher amounts of delphinidin 3-glucoside (Dp-3-glc), malvidin 3-glucoside (Mv-3-glc), and malvidin 3-galactoside (Mv-3-gal). Dp-3-glc was 263.8 (mg/100 g) in Saskatoon berry and 84.4 (mg/100 g) in wild blueberry, whereas the corresponding values for Mv-3-glc in these berries were 47.4 and 139.6 (mg/100 g), respectively. Raspberry, strawberry, and chokecherry contained higher amounts of cyanidin 3-glucoside (Cy-3-glc), cyanidin 3-rutinoside (Cy-3-rut), and pelargonidin 3-glucoside (Pg-3-glc). The total anthocyanin content of Manitoba fruits followed the order: Saskatoon berry and blueberry (high anthocyanin berries), raspberry and chokecherry (medium anthocyanin berries), strawberry (low anthocyanin berries), and seabuckthorn (negligible anthocyanin berries). This study demonstrated that Saskatoon berries and wild blueberries have high potential value for fruit growers as well as the food and nutraceutical manufacturers because of their high anthocyanin contents.     

常规稻与杂交稻谷的仿生电子鼻分类识别

气味是进行稻谷品种及其品质识别的重要方法之一,作为一种基于仿生嗅觉的机器检测方法,仿生电子鼻在水稻品种的分类识别中具有较好的应用前景。常规稻与杂交稻在食味品质等方面存在一定的差异,为了解应用电子鼻进行常规稻谷与杂交稻谷识别的可行性,采用PEN3电子鼻对同季同地域收获的3种常规稻(中香1号、湘晚13、瑶平香)和3种杂交稻(伍丰优T025、品36、优优122)稻谷样品的气味信息进行了采集和分析。首先通过过载分析(Loadings)法分析了电子鼻检测稻谷气体挥发物时的各传感器贡献率,分别针对基于特征值的提取和稻谷气味检测对电子鼻传感器阵列中的传感器进行了优选,阐明了稻谷气体挥发物检测中应以对硫化物、氮氧化合物、芳香成分和有机硫化物敏感的传感器为主。随后,分别采用主成分分析法(principal component analysis,PCA)、线性判别法(linear discriminant analysis,LDA)和BP神经网络对6种不同稻谷之间、常规稻与杂交稻之间的分类识别进行了研究。结果表明,PCA分析法与LDA分析法在对6种不同稻谷之间的分类以及常规稻与杂交稻之间的分类中均未取得理想的效果,存在部分样本数据点重叠或样本数据点较近的情况,在实际应用中易发生混淆;而BP神经网络在对6种不同稻谷之间的分类中对测试集的识别正确率分别达到了90%,在常规稻与杂交稻之间的分类识别中对测试集的识别正确率达到了96.7%。上述试验验证了电子鼻用于常规稻与杂交稻稻谷分类识别的有效性,为常规稻与杂交稻的快速、无损分类识别提供了一种新的方法。     

Classification of gilthead sea bream (Sparus aurata) from 1H NMR lipid profiling combined with principal component and linear discriminant analysis

The combination of (1)H NMR fingerprinting of lipids from gilthead sea bream (Sparus aurata) with nonsupervised and supervised multivariate analysis was applied to differentiate wild and farmed fish and to classify farmed specimen according to their areas of production belonging to the Mediterranean basin. Principal component analysis (PCA) applied on processed (1)H NMR profiles made a clear distinction between wild and farmed samples. Linear discriminant analysis (LDA) allowed classification of samples according to the geographic origin, as well as for the wild and farmed status using both PCA scores and NMR data as variables. Variable selection for LDA was achieved with forward selection (stepwise) with a predefined 5% error level. The methods allowed the classification of 100% of the samples according to their wild and farmed status and 85-97% to geographic origin. Probabilistic neural network (PNN) analyses provided complementary means for the successful discrimination among classes investigated.     

基于高光谱图像处理的大豆品种识别(英文)

大豆组分(油,蛋白质,脂肪等)在不同的大豆品种间差异很大。对于提高大豆品质来说,大豆品种识别是一个关键因素。该文利用高光谱图像技术对不同的大豆品种进行识别。利用高光谱成像系统获取大豆样本1 000～2 500 nm范围的光谱反射数据;应用主成分分析法(PCA,principal component analysis)对获取到的光谱数据进行数据降维并去除冗余数据;在分类算法中将得分高的主成分值作为输入特征,通过PCA方法从每个特征图像中提取4个特征变量(能量、熵、惯性矩和相关性);对于具体特征提取,从16个特征变量中提取8个重要特征参数;根据选择的特征,应用神经网络方法构建分类器;训练精度精度达到97.50%,平均测试精度达到93.88%以上。结果表明,应用高光谱图像技术结合神将网络建模方法可以对大豆品种进行分类。     

Detecting Vitreous Wheat Kernels Using Reflectance and Transmittance Image Analysis

The proportion of vitreous durum kernels in a sample is an important grading attribute in assessing the quality of durum wheat. The current standard method of determining wheat vitreousness is performed by visual inspection, which can be tedious and subjective. The objective of this study was to evaluate an automated machine‐vision inspection system to detect wheat vitreousness using reflectance and transmittance images. Two subclasses of durum wheat were investigated in this study: hard and vitreous of amber color (HVAC) and not hard and vitreous of amber color (NHVAC). A total of 4,907 kernels in the calibration set and 4,407 kernels in the validation set were imaged using a Cervitec 1625 grain inspection system. Classification models were developed with stepwise discriminant analysis and an artificial neural network (ANN). A discriminant model correctly classified 94.9% of the HVAC and 91.0% of the NHVAC in the calibration set, and 92.4% of the HVAC and 92.7% of the NHVAC in the validation set. The classification results using the ANN were not as good as with the discriminant methods, but the ANN only used features from reflectance images. Among all the kernels, mottled kernels were the most difficult to classify. Both reflectance and transmittance images were helpful in classification. In conclusion, the Cervitec 1625 automated visionbased wheat quality inspection system may provide the grain industry with a rapid, objective, and accurate method to determine the vitreousness of durum wheat.     

黄花梨果形的机器视觉识别方法研究

黄花梨的果形是分级的重要特征之一。利用机器视觉采集黄花梨图像,研究了不规则果品的形状描述方法,提出在黄花梨的分级过程中采用傅立叶变换与傅立叶反变换对来描述果形,开发了基于人工神经网络的果形识别软件。研究发现该傅立叶描述子的前１６个谐波的变化特性足以代表梨体的主要形状,采用傅立叶描述子与人工神经网络相结合的方法进行果形识别的精确率可达９０％。而且只要有合适的训练对,该方法也可以用来对其它水果进行外形识别     

Detection and quantification of apple adulteration in diluted and sulfited strawberry and raspberry purées using visible and near-infrared spectroscopy

Adulteration of sulfited strawberry and raspberry purées by apple is a commercial problem. Strawberry (n = 31) and raspberry (n = 30) purées were prepared from Irish-grown fruit and adulterated at levels of 10-75% w/w using cooking apples. Visible and near-infrared transflectance spectra were recorded using a 0.1 mm sample thickness. Classification and quantification models were developed using raw and scatter-corrected and/or derivatized spectral data. Classification as pure strawberry or raspberry was attempted using soft independent modeling of class analogy. The best models used spectral data in the wavelength ranges 400-1098 nm (strawberry) and 750-1098 nm (raspberry) and produced total correct classification rates of 75% (strawberry) and 95% (raspberry). Quantification of apple content was performed using partial least-squares regression. Lowest predictive errors obtained were 11.3% (raspberry) and 9.0% (strawberry). These results were obtained using spectral data in the wavelength ranges 400-1880 and 1100-1880 nm, respectively. These results suggest minimum detection levels of apple in soft fruit purées of approximately 25 and 20% w/w for raspberry and strawberry, respectively.     

Different polyphenolic components of soft fruits inhibit alpha-amylase and alpha-glucosidase

Polyphenol-rich extracts from soft fruits were tested for their ability to inhibit alpha-amylase and alpha-glucosidase. All extracts tested caused some inhibition of alpha-amylase, but there was a 10-fold difference between the least and most effective extracts. Strawberry and raspberry extracts were more effective alpha-amylase inhibitors than blueberry, blackcurrant, or red cabbage. Conversely, alpha-glucosidase was more readily inhibited by blueberry and blackcurrant extracts. The extent of inhibition of alpha-glucosidase was related to their anthocyanin content. For example, blueberry and blackcurrant extracts, which have the highest anthocyanin content, were the most effective inhibitors of alpha-glucosidase. The extracts most effective in inhibiting alpha-amylase (strawberry and raspberry) contain appreciable amounts of soluble tannins. Other tannin-rich extracts (red grape, red wine, and green tea) were also effective inhibitors of alpha-amylase. Indeed, removing tannins from strawberry extracts with gelatin also removed inhibition. Fractionation of raspberry extracts on Sephadex LH-20 produced an unbound fraction enriched in anthocyanins and a bound fraction enriched in tannin-like polyphenols. The unbound anthocyanin-enriched fraction was more effective against alpha-glucosidase than the original extract, whereas the alpha-amylase inhibitors were concentrated in the bound fraction. The LH-20 bound sample was separated by preparative HPLC, and fractions were assayed for inhibition of alpha-amylase. The inhibitory components were identified as ellagitannins using LC-MS-MS. This study suggests that different polyphenolic components of fruits may influence different steps in starch digestion in a synergistic manner.     

近红外光谱结合小波变换预处理在蜂蜜产地快速检测中的应用

应用近红外光谱结合化学计量学方法对蜂蜜产地进行了判别分析。kennard-Stone法划分训练集和预测集。光谱用一阶导数加自归一化预处理后,再用小波变换（WT）进行压缩和滤噪。结合滤波后光谱信息,分别用径向基神经网络（RBFNN）和偏最小二乘-线性判别分析（PLS-LDA）建立了苹果蜜产地和油菜蜜产地判别模型。对不同小波基和分解尺度进行了详细讨论。对苹果蜜,WT-RBFNN模型和WT-PLS-LDA模型都是小波基为db1、分解尺度为2时的预测精度最好,都为96.2%。对油菜蜜：WT-RBFNN模型在小波基为db4和分解尺度为1时,预测精度最好;WT-PLS-LDA模型在小波基为db9、分解尺度也为1时,预测精度最好,为90.5%;预测精度WT-PLS-LDA模型优于WT-RBFNN模型。研究表明：WT结合线性的PLS-LDA建模比WT结合非线性的RBFNN建模更适于蜂蜜产地鉴别;近红外光谱结合WT-PLS-LDA可实现对蜂蜜产地的快速无损检测,为蜂蜜产地鉴别提供了一种新方法。     

Phenotypic diversity and delimitation between wild and cultivated forms of the genus <Emphasis Type="Italic">Pyrus</Emphasis> in North-eastern Spain based on morphometric analyses

We investigated morphological evidence that might allow wild Pyrus spp. to be distinguished from cultivated material (Pyrus communis L.) in the North-eastern Iberian Peninsula. 134 pear trees were identified in the wild and characterized by 13 quantitative and 13 qualitative leaf–shoot and fruit traits. The trees were visually classified into two preliminary groups of wild and cultivated material and discriminant functions, based on a reference collection for allocating individuals to one of the groups, were constructed. Both classifications were compared with a near-optimal numerical classification (the two-stage Ward-MLM strategy) using two criteria. The visual assignment of trees allocated 60% of trees to the wild group and 40% to the cultivated group. The overall discrepancy rate between the field classification and the discriminant analysis was low (17.4%). In general, wild individuals had smaller leaves, shorter petioles and more rounded and smaller fruits than their cultivated counterparts. They also had small-to-intermediate petiole widths, thorns on their shoots and straight or convex fruit profiles. However, the Ward-MLM strategy always formed better groups, in terms of the two criteria used, in all the continuous and categorical variables, for both leaf–shoot and fruit traits. Likewise, the agreement between classifications (discriminant analysis and Ward-MLM strategy) was only partial, with some Ward-MLM groups composed of both wild and cultivated material in similar proportions. This result suggests a limited success in identifying genuine wild individuals based on morphometric data, which can be ascribed either to poor phenotypic diversity and lack of distinguishing traits among species or to widespread crossability and subsequent development of hybrid/introgressant populations between wild and cultivated specimens.     

基于近红外光谱和深度学习数据增强的大米品种检测

近红外光谱和深度学习结合的思路是大米品种检测的重要研究方向，其准确检测模型的建立依赖大规模的样本数据，然而采集和预处理样本耗时巨大，对准确性的提升造成限制。为解决上述不足，便于深入探究近红外光谱结合深度学习方法在大米品种检测领域应用的可行性，该研究提出基于近红外光谱结合改进型深度卷积生成式对抗神经网络（deep convolutional generative adversarial network，DCGAN）数据增强的大米品种检测方法。首先，在相同环境下采集4种大米品种的近红外光谱并对原始光谱数据进行预处理，使用去趋势校正（detrend correction，DC）和无信息变量消除算法（uninformative variable elimination，UVE）消除无用光谱特征点。然后，建立改进型DCGAN模型对预处理后的光谱数据进行数据增强，对比试验结果表明，改进型DCGAN相比与传统数据增强方法，改进型DCGAN生成数据的结构相似度指标更优。最后，研究不同数据增强方法结合不同分类方法建立大米品种分类模型的性能，对比试验结果表明，改进型DCGAN数据增强结合一维卷积神经网络（one-dimensional convolution neural network，1D-CNN）分类算法所建模型面向测试集的准确率最高，为98.21%，为简便准确的大米品种检测方案提供了新思路。     

水稻重金属污染胁迫光谱分析模型的区域应用与验证

根据样地试验建立的农作物重金属污染胁迫光谱分析模型通过卫星遥感数据进行大尺度区域应用是农作物重金属污染遥感评价必须解决的关键问题。该文以吉林长春市3块重金属污染程度不同的水稻农田样地为试验区,采集水稻冠层ASD(Analytical Spectral Devices)数据、叶片叶绿素含量和土壤重金属含量,并获取准同步的Hyperion数据,通过多元逐步回归分析筛选与重金属污染胁迫响应敏感的光谱指数,并运用BP人工神经网络模型构建其与表征重金属污染胁迫程度的叶绿素含量的数学关系模型。结果表明,样地水稻重金属污染胁迫光谱分析模型中的BP网络结构为4-11-7-1、传递函数为logsig,其对各类污染胁迫水平的判别精度均为100%;将所建立的样地水稻重金属污染胁迫光谱分析模型通过Hyperion影像,进行大面积推广并验证,得到其对各类污染胁迫水平的判别精度均超过80%。该研究为样地水稻重金属污染胁迫光谱分析模型的大面积推广应用提供了借鉴意义。     

近红外光谱结合化学计量学方法检测蜂蜜产地

为了实现蜂蜜产地的快速判别,应用近红外光谱结合化学计量学方法对蜂蜜产地进行了判别分析。kennard-Stone法划分训练集和预测集。光谱用一阶导数加自归一化预处理后,再用小波变换（WT）进行压缩和滤噪。结合滤波后光谱信息,分别用径向基神经网络（RBFNN）和偏最小二乘-线性判别分析（PLS-LDA）建立了苹果蜜产地和油菜蜜产地的判别模型。对不同小波基和分解尺度进行了讨论。对苹果蜜,WT-RBFNN模型和WT-PLS-LDA模型都是小波基为db1、分解尺度为2时的预测精度较好,都为96.2%。对油菜蜜：WT-RBFNN模型在小波基为db4和分解尺度为1时,预测精度较好,为85.7%;WT-PLS-LDA模型在小波基为db9、分解尺度也为1时,预测精度较好,为90.5%。研究表明：WT结合线性的PLS-LDA建模比WT结合非线性的RBFNN建模更适于蜂蜜产地判别;近红外光谱技术具有快速判别蜂蜜产地的潜力。     

贵州名优茶与产地土壤矿质元素相关性及其溯源影响的研究

为探究贵州省名优茶产地不同深层土壤对茶叶矿质元素溯源效果的影响,以土壤-茶叶的多矿质元素法结合主成分分析(PCA)、反向传播(BP)神经网络法、逐步线性判别分析(SLDA)对茶叶产地进行溯源。结果显示,不同产地的茶叶和土壤具有独特的矿质元素指纹,茶叶中Fe、Mn、K、Ca、Mg、Cu的含量与土壤中对应元素含量显著相关(P<0.05),以这6种矿质元素含量结合PCA可有效区分茶叶的地理起源;不同深层土壤对茶叶产地的溯源有不同影响,通过SLDA法、BP神经网络法明确了以60~80 cm的土层进行产地溯源的效果最优,产地溯源验证判别率分别为98.5%(SLDA法)和100%(BP神经网络法),并基于SLDA法确定了Zn、Cu、P、Mn、Fe、Mg和K 7种元素构建的贵州名优茶产地溯源模型。此外,研究发现茶叶产地溯源几乎不受茶树品种的影响。本研究结果为贵州省名优茶矿质元素指纹图谱的建设及土壤-茶叶产地溯源的关系研究提供了理论依据。