激光诱导击穿光谱结合CARS变量选择方法定量检测奶粉中的镁

为监测奶粉中的镁(Mg)元素含量,本研究利用激光诱导击穿光谱(LIBS)技术对奶粉中Mg元素进行定量检测。对于每个样品,采用压片机在20 MPa压力下进行压片处理,然后利用高精度光谱仪在200~750 nm波段范围内获取压片样品的LIBS光谱。根据LIBS光谱特征,将光谱划分为4个波段,并进行初步的波段优选和光谱预处理分析。在此基础上,采用竞争性自适应重加权算法(CARS)对波长变量进行优选,再应用偏最小二乘法(PLS)建立奶粉中Mg元素含量的预测模型,并对预测集样本进行预测。研究结果表明,LIBS技术结合CARS变量选择方法可以用于奶粉中Mg元素含量的定量检测,最优CARS-PLS预测模型的校正集和预测集的决定系数及平均相对误差分别为0.9999,0.20%和0.9742,3.29%,优于原始光谱所建立的PLS模型,且所用波长变量仅为PLS模型的7.7%。由此表明,CARS方法能有效选择有用的波长变量,可简化预测模型及提高预测模型的稳定性。本研究为奶粉中镁元素含量的快速定量分析提供参考。     

基于高光谱图像的生菜叶片氮素含量预测模型研究

为了便于更经济合理地为作物施肥,建立一种无损检测作物氮营养元素的高光谱图像模型。本实验以生菜为研究对象,无土栽培各氮素水平的生菜叶样本,在莲座期,采集生菜叶片样本的高光谱图像(390~1050 nm),同时采用凯氏定氮法测定对应生菜叶片样本的全氮含量。通过ENVI软件提取出生菜叶片中感兴趣区域的平均光谱作为该样本原始光谱信息,分别使用平滑处理(Smoothing)、多元散射矫正(MSC)、标准正态变量变换结合去趋势(SNV detrending)、一阶导数法(First derivative)、二阶导数法(Second derivative)、正交信号矫正(OSC)等预处理方法对样本原始光谱进行处理,然后利用偏最小二乘回归法(Partial least squares regression,PLSR)分别建立样本全波段光谱信息与氮含量的关系模型,研究各预处理方法对氮含量模型的影响,结果表明,使用OSC预处理的模型效果最好。为了简化模型,根据OSC预处理光谱后的模型的PLSR回归系数优选出敏感波长,利用训练集中样本的敏感波长光谱信息与氮含量数据重新构建PLSR回归模型,并利用测试集样本进行测试试验。结果表明,该模型得到校正集和预测集的决定系数(R2p)分别为0.89,0.81;均方根误差RMSEC,RMSEP分别为0.33,0.45。该回归模型大大降低了自变量个数,简化了模型,并且取得了较优的效果,这为生菜氮素含量预测提供了一种新的快速有效方法。     

可见-近红外光谱结合CARS优化模型预测棕榈油碘值

采用可见-近红外透射光谱结合CARS变量优选方法优化模型,对棕榈油碘值进行近红外定量分析。通过将使用不同预处理方法产生的建模效果进行比较,找到了理想的预处理方法,通过CARS变量选择方法优选出与棕榈油碘值相关的有效波点共60个,利用60个有效波点建立棕榈油碘值优化模型。根据优化模型的建模集相关系数(R_c=0.9814)和预测集相关系数(R_p=0.9806),得到的建模均方根误差(RM SEC=0.0398)和预测均方根误差(RM SEP=0.0406)优于采用全波段建立的模型得到的系数误差。利用可见近红外透射光谱结合CARS变量优选方法,简化了棕榈油碘值模型,并能够保证碘值预测的准确度。     

苹果产地差异对可溶性固形物近红外光谱检测模型影响的研究

为更好地利用近红外光谱预测苹果可溶性固形物含量,减少产地差异对近红外光谱检测模型的影响,以4种不同产地的富士苹果为研究对象,采用基于x-y共生距离的样本划分方法分别对不同产地的苹果选取代表性样本作为校正集,利用偏最小二乘算法,建立和比较单一产地和混合产地下的苹果可溶性固形物近红外光谱检测模型,并结合竞争性自适应重加权算法(CARS)和连续投影算法(SPA)对苹果可溶性固形物的建模变量进行筛选。相比单一产地和其它混合产地模型,混合所有4种苹果产地的校正集样本建立的模型取得了最好的预测结果,另外,结合CARS-SPA筛选的16个特征波长,模型得到了进一步简化,其预测相关系数和预测均方根误差分别为0.978和0.441°Brix。结果表明,利用多个产地的苹果样本建立的混合模型,结合有效特征波长,可提高对苹果可溶性固形物含量的预测精度,减小产地差异对可溶性固形物近红外光谱检测的影响。     

基于人血浆荧光光谱与CARS-PLS-LDA的结直肠癌早期检测

该文利用竞争性自适应加权算法（CARS）筛选重要的人血浆荧光光谱变量,并结合偏最小二乘法判别分析（PLS－LDA）建立了结直肠癌患者与非癌患者的分类模型,同时与全波长模型和基于平行因子分析（PARAFAC）建立的模型进行比较。从模型评价指标看,CARS－PLS－LDA的性能显著优于全波长模型和基于PARAFAC的模型。高波未稀释组和低波稀释组的荧光光谱结合CARS－PLS－LDA分类模型的AUC（Area under curve）值均高于0.9,可有效地识别结直肠癌患者。结果表明,CARS变量筛选能够明显改善结直肠癌分类模型的性能,有助于后续癌症临床诊断工具的开发与研究。     

激光诱导击穿光谱结合CARS变量选择方法定量检测倍硫磷含量

利用双脉冲激光诱导击穿光谱(LIBS)技术对溶液中的倍硫磷含量进行定量检测。采用二通道高精度光谱仪采集不同浓度倍硫磷样品在206.28~481.77 nm波段的LIBS光谱,并对光谱进行多元散射校正(MSC)、标准正态变量变换(SNV)及3点平滑预处理,根据偏最小二乘(PLS)建模确定最优的预处理方法。在此基础上,利用竞争性自适应重加权算法(CARS)筛选与倍硫磷相关的重要变量,然后应用PLS回归建立溶液中倍硫磷含量的定量分析模型,并与单变量定量分析模型及未变量选择的PLS定量分析模型进行比较。结果表明,相比单变量定量分析模型及原始光谱PLS定量分析模型,CARS-PLS定量分析模型的性能更优,其模型的校正集和预测集的决定系数及平均相对误差分别为0.969 4、15.537%和0.995 9、5.016%。此外,与原始光谱PLS模型相比,CARS-PLS模型仅使用其中1.9%的波长变量,但预测集平均误差却由9.829%下降为5.016%。由此可见,LIBS技术检测溶液中的倍硫磷含量具有一定的可行性,且CARS方法能简化定量分析模型,提高模型的预测精度。     

基于多阶导数拉曼光谱组合技术的矿物油模式分类

为了实现对法庭科学领域重质矿物油物证的快速、准确、无损的鉴定,该文基于光谱分析技术提出了一种多阶导数光谱数据组合分析的方法。收集了80种不同型号、不同厂家的重质矿物油样本,利用傅里叶变换拉曼光谱分析法采集样本的原始光谱数据和导数光谱数据,并通过结合化学计量学构建分类模型。在构建的主成分分析(PCA)结合径向基函数神经网络(RBF)分类模型中,对单独的原始光谱、一阶导数谱和二阶导数谱数据的训练集准确率分别为80.0%、86.7%和86.2%,测试集准确率分别为73.3%、80.0%和72.7%;对组合后的原始光谱+一阶导数谱、原始光谱+二阶导数谱和一阶导数谱+二阶导数谱数据的分类中,训练集准确率分别为97.0%、96.7%和100%,测试集准确率分别为85.7%、90.0%和100%。结果表明,对组合后的导数光谱与原始光谱构建分类模型,准确率更高。其中,基于一阶导数谱+二阶导数谱数据构建的PCA结合RBF分类模型的结果最为理想,准确率达100%。而K最近邻算法模型由于受到样本不均匀的影响,整体分类准确率均较低。利用组合的导数光谱与原始光谱数据构建分类模型能够实现对重质矿物油样本的快速、准确、无损鉴别,可为光谱组合技术在法庭科学及其他分析测试领域的应用提供一定的借鉴和参考。     

基于小波变换-遗传算法-偏最小二乘的草莓糖度检测研究

采用近红外漫反射光谱分析技术,对草莓糖度进行了无损检测研究。利用便携式近红外光谱仪采集草莓样品在600～1 100 nm波段内的漫反射光谱数据。首先利用小波变换(WT)多分辨率方法对光谱数据进行去噪预处理,然后利用遗传算法(GA)优选特征波长,最后运用偏最小二乘法(PLS)建立草莓糖度的WT-GA-PLS校正模型。该模型校正集的相关系数R_C为0.9395,校正集的均方根误差RMSEC为0.1615,预测集的相关系数R_P为0.9652,预测集的均方根误差EMSEP为0.5042。与全光谱模型(FS-PLS)和小波变换模型(WT-PLS)相比,该模型预测能力更强,稳健性更优。     

基于近红外光谱的食用植物油中反式脂肪酸含量快速定量检测及模型优化研究

利用近红外光谱技术对食用植物油中反式脂肪酸(Trans fatty acids,TFA)含量进行快速定量检测,并通过波段选择、预处理方法、变量筛选及建模方法对TFA含量预测模型进行优化.采用AntarisⅡ傅里叶变换近红外光谱仪在4000~10000 cm-1光谱范围采集98个食用植物油样本的近红外透射光谱,然后采用气相色谱法测定TFA的真实含量.首先,对样本原始光谱进行波段、预处理方法优选;在此基础上,采用竞争自适应重加权法(Competitive adaptive reweighted sampling,CARS)筛选TFA相关的重要变量,最后应用主成分回归、偏最小二乘和最小二乘支持向量机方法分别建立食用植物油中TFA含量的预测模型.研究结果表明,近红外光谱技术检测食用植物油中的TFA含量是可行的,优化后的最佳预测模型的校正集和预测集R2分别为0.992和0.989,RMSEC和RMSEP分别为0.071%和0.075%.最佳预测模型所用的变量仅26个,占全波段变量的0.854%.此外,与全波段偏最小二乘预测模型相比,其预测集R2由0.904上升为0.989,RMSEP由0.230%下降为0.075%.由此表明,模型优化非常必要,CARS能有效筛选TFA相关的重要变量,极大减少建模变量数,从而简化预测模型,并较大提高预测模型的精度和稳定性.     

基于卷积神经网络和近红外光谱的茶叶品种和等级鉴别

以5个品种茶叶和4个不同等级龙井茶叶为研究对象,利用近红外光谱与卷积神经网络相结合的方法,实现茶叶品种和等级的鉴别。对实验采集得到的800～2 500 nm原始光谱使用小波分析(WT)算法进行预处理,对预处理后的光谱数据分别采用联合区间偏最小二乘法(siPLS)、连续投影算法(SPA)、竞争性自适应重加权算法(CARS)提取特征波长,然后建立卷积神经网络(CNN)分类模型,实现茶叶品种和等级的鉴别。结果显示：SPA+CNN模型对品种和等级鉴别的准确率分别达到了95.83%和96.67%,CARS+CNN将准确率进一步提升到97.72%和98.67%。最后使用平移法、线性叠加法、添加噪声法对光谱数据集进行数据增强,验证卷积神经网络模型的稳定性。研究结果表明,特征波长提取结合卷积神经网络,可以实现对茶叶品种和等级的无损鉴别。为后续开发动态在线检测设备提供了高效、无损、快速的技术支持。     

Key wavelengths screening using competitive adaptive reweighted sampling method for multivariate calibration

By employing the simple but effective principle ‘survival of the fittest’ on which Darwin's Evolution Theory is based, a novel strategy for selecting an optimal combination of key wavelengths of multi-component spectral data, named competitive adaptive reweighted sampling (CARS), is developed. Key wavelengths are defined as the wavelengths with large absolute coefficients in a multivariate linear regression model, such as partial least squares (PLS). In the present work, the absolute values of regression coefficients of PLS model are used as an index for evaluating the importance of each wavelength. Then, based on the importance level of each wavelength, CARS sequentially selects N subsets of wavelengths from N Monte Carlo (MC) sampling runs in an iterative and competitive manner. In each sampling run, a fixed ratio (e.g. 80%) of samples is first randomly selected to establish a calibration model. Next, based on the regression coefficients, a two-step procedure including exponentially decreasing function (EDF) based enforced wavelength selection and adaptive reweighted sampling (ARS) based competitive wavelength selection is adopted to select the key wavelengths. Finally, cross validation (CV) is applied to choose the subset with the lowest root mean square error of CV (RMSECV). The performance of the proposed procedure is evaluated using one simulated dataset together with one near infrared dataset of two properties. The results reveal an outstanding characteristic of CARS that it can usually locate an optimal combination of some key wavelengths which are interpretable to the chemical property of interest. Additionally, our study shows that better prediction is obtained by CARS when compared to full spectrum PLS modeling, Monte Carlo uninformative variable elimination (MC-UVE) and moving window partial least squares regression (MWPLSR).     

Rapid prediction of moisture content of dehydrated prawns using online hyperspectral imaging system

Because the shape of prawn is not round, spectroscopy instruments cannot measure the spectra of the whole prawn without containing background information. In this study, an online hyperspectral imaging system in the spectral region of 380–1100 nm was developed to determine the moisture content of prawns at different dehydrated levels. Hyperspectral images of prawns were acquired at different dehydration periods. The spectra of prawns then were extracted from hyperspectral images based on ‘Manual Prawn Mask’ and ‘Automatic Prawn Mask’, respectively. Spectral data were analyzed using partial least squares regression (PLSR) and least-squares support vector machines (LS-SVM) to establish the calibration models, respectively. Successive projections algorithm (SPA) was first applied for the optimal wavelength selection in the hyperspectral image analysis. Out of 482 wavelengths, only twelve wavelengths (428, 445, 544, 569, 629, 672, 697, 760, 827, 917, 958, and 999 nm) were selected by SPA as the optimum wavelengths for moisture prediction. Based on these optimum wavelengths, a multiple linear regression (MLR) calibration model was established and used to obtain the moisture distribution of each prawn. The overall results of this study revealed the potentiality of hyperspectral imaging as an objective and non-destructive method to obtain the content and distribution of moisture of prawns whose shapes are not round.     

Rapid Determination of Cellulose and Hemicellulose Contents in Corn Stover Using Near-Infrared Spectroscopy Combined with Wavelength Selection

The contents of cellulose and hemicellulose (C and H) in corn stover (CS) have an important influence on its biochemical transformation and utilization. To rapidly detect the C and H contents in CS by near-infrared spectroscopy (NIRS), the characteristic wavelength selection algorithms of backward partial least squares (BIPLS), competitive adaptive reweighted sampling (CARS), BIPLS combined with CARS, BIPLS combined with a genetic simulated annealing algorithm (GSA), and CARS combined with a GSA were used to select the wavelength variables (WVs) for C and H, and the corresponding regression correction models were established. The results showed that five wavelength selection algorithms could effectively eliminate irrelevant redundant WVs, and their modeling performance was significantly superior to that of the full spectrum. Through comparison and analysis, it was found that CARS combined with GSA had the best comprehensive performance; the predictive root mean squared errors of the C and H regression model were 0.786% and 0.893%, and the residual predictive deviations were 3.815 and 12.435, respectively. The wavelength selection algorithm could effectively improve the accuracy of the quantitative analysis of C and H contents in CS by NIRS, providing theoretical support for the research and development of related online detection equipment.     

基于高光谱成像结合分光光度技术的喷墨打印墨水种类鉴别方法

在法庭科学实践中,往往需要通过对文件中字迹墨水的成分分析来精确地判定检材和样本文件的同一性。利用高光谱成像和分光光度技术结合化学计量法,提出了一种对喷墨打印墨水分类的方法。采集14台不同品牌、型号的四色喷墨打印墨水高光谱数据和色度值。计算出平均色度值后进行PCA降维处理和K-Means聚类分析,将样品初步分类。之后应用LightGBM模型、XGBoost模型和SVM模型共三种分类模型,以1:4的比例确定测试集和训练集,对聚类分析结果中每一类别的样品进行逐一鉴别。结果表明,LightGBM和XGBoost对四色样品的分类精度都能达到95%以上,SVM的分类精度为100%。提出的方法能够做到无损、准确、快速地将不同品牌乃至型号的喷墨打印墨水进行区分。     

Wavelength selection framework for classifying food and pharmaceutical samples into multiple classes

Near infrared (NIR) spectroscopy is an efficient, low‐cost analytical technique widely applied to identify the origin of food and pharmaceutical products. NIR spectra‐based classification strategies typically use thousands of equally spaced wavelengths as input information, some of which may not carry relevant information for product classification. When that is the case, the performance of predictive and exploratory multivariate techniques may be undermined by such noisy information. In this paper, we propose an iterative framework for selecting subsets of NIR wavelengths aimed at classifying samples into categories. For that matter, we integrate Principal Components Analysis (PCA) and three classification techniques: k‐Nearest Neighbor (KNN), Probabilistic Neural Network (PNN) and Linear Discriminant Analysis (LDA). PCA is first applied to NIR data, and a wavelength importance index is derived based on the PCA loadings. Samples are then categorized using the wavelength with the highest index and the classification accuracy is calculated; next, the wavelength with the second highest index is inserted into the dataset and a new classification is performed. This forward‐based iterative procedure is carried out until all original wavelengths are inserted into the dataset used for classification. The subset of wavelengths leading to the maximum accuracy is chosen as the recommended subset. Our propositions performed remarkably well when applied to four datasets related to food and pharmaceutical products. Copyright © 2016 John Wiley & Sons, Ltd.     

Rapid detection of Escherichia coli contamination in packaged fresh spinach using hyperspectral imaging

A rapid method based on hyperspectral imaging for detection of Escherichia coli contamination in fresh vegetable was developed. E. coli K12 was inoculated into spinach with different initial concentrations. Samples were analyzed using a colony count and a hyperspectroscopic technique. A hyperspectral camera of 400-1000 nm, with a spectral resolution of 5 nm was employed to acquire hyperspectral images of packaged spinach. Reflectance spectra were obtained from various positions on the sample surface and pretreated using Sawitzky-Golay. Chemometrics including principal component analysis (PCA) and artificial neural network (ANN) were then used to analyze the pre-processed data. The PCA was implemented to remove redundant information of the hyperspectral data. The ANN was trained using Bayesian regularization and was capable of correlating hyperspectral data with number of E. coli. Once trained, the ANN was also used to construct a prediction map of all pixel spectra of an image to display the number of E. coli in the sample. The prediction map allowed a rapid and easy interpretation of the hyperspectral data. The results suggested that incorporation of hyperspectral imaging with chemometrics provided a rapid and innovative approach for the detection of E. coli contamination in packaged fresh spinach.     

Detection of trace amounts of Ni by laser-induced fluorescence in graphite furnace with intensified charge coupled device

A laser-induced fluorescence in graphite furnace (LIF-GF) set-up has been equipped with an intensified CCD detector (ICCD) that enables simultaneous multichannel detection of large wavelength regions. The main advantages of such a system in comparison with conventional photomultiplier detection are: simultaneous detection of several fluorescence wavelengths for easy characterization of excitation and fluorescence spectra and for an increase of the absolute sensitivity and spectral selectivity; simultaneous monitoring of background signals, such as those due to matrix interferences, blackbody radiation and scattered laser light; decrease of the susceptibility to radio-frequency pick-ups emitted from the pump laser due to the delayed read-out procedure; time-resolved studies of fluorescence spectra for improved elemental selectivity or for studies of atomization processes, and a possibility to perform two-dimensional imaging of height distributions of atomization and, in combination with an imaging spectrometer, diffusion processes in the furnace. The first work on LIF-GF with ICCD detection has been performed on Ni. The most sensitive and versatile excitation and detection wavelengths have been identified. Detection limits in water solutions by the LIF-GF technique have been improved by two orders of magnitude and are found to be 0.015 pg with ICCD and 0.01 pg using a photomultiplier at the most sensitive excitation and detection wavelengths. Nickel in concentrations has been detected in aqueous standard reference samples with sodium concentrations ranging from to % (riverine water and estuarine water) with good accuracy and precision. The Ni contents of standard riverine and estuarine water were determined to 1.00 ± 0.11 and 0.75 ± 0.07 ng/ml, respectively. The certified concentrations are 1.03 ± 0.10 and 0.743 ± 0.078 .     

The Classification of Rice Blast Resistant Seed Based on Ranman Spectroscopy and SVM

Rice blast is a serious threat to rice yield. Breeding disease-resistant varieties is one of the most economical and effective ways to prevent damage from rice blast. The traditional identification of resistant rice seeds has some shortcoming, such as long possession time, high cost and complex operation. The purpose of this study was to develop an optimal prediction model for determining resistant rice seeds using Ranman spectroscopy. First, the support vector machine (SVM), BP neural network (BP) and probabilistic neural network (PNN) models were initially established on the original spectral data. Second, due to the recognition accuracy of the Raw-SVM model, the running time was fast. The support vector machine model was selected for optimization, and four improved support vector machine models (ABC-SVM (artificial bee colony algorithm, ABC), IABC-SVM (improving the artificial bee colony algorithm, IABC), GSA-SVM (gravity search algorithm, GSA) and GWO-SVM (gray wolf algorithm, GWO)) were used to identify resistant rice seeds. The difference in modeling accuracy and running time between the improved support vector machine model established in feature wavelengths and full wavelengths (200–3202 cm⁻¹) was compared. Finally, five spectral preproccessing algorithms, Savitzky–Golay 1-Der (SGD), Savitzky–Golay Smoothing (SGS), baseline (Base), multivariate scatter correction (MSC) and standard normal variable (SNV), were used to preprocess the original spectra. The random forest algorithm (RF) was used to extract the characteristic wavelengths. After different spectral preproccessing algorithms and the RF feature extraction, the improved support vector machine models were established. The results show that the recognition accuracy of the optimal IABC-SVM model based on the original data was 71%. Among the five spectral preproccessing algorithms, the SNV algorithm’s accuracy was the best. The accuracy of the test set in the IABC-SVM model was 100%, and the running time was 13 s. After SNV algorithms and the RF feature extraction, the classification accuracy of the IABC-SVM model did not decrease, and the running time was shortened to 9 s. This demonstrates the feasibility and effectiveness of IABC in SVM parameter optimization, with higher prediction accuracy and better stability. Therefore, the improved support vector machine model based on Ranman spectroscopy can be applied to the fast and non-destructive identification of resistant rice seeds.     

A probability-based multivariate statistical algorithm for autofluorescence spectroscopic identification of oral carcinogenesis

A probability-based multivariate statistical algorithm combining partial least-squares (PLS) and logistic regression was developed to identify the development stages of oral cancer through analysis of autofluorescence spectra of oral tissues. Tissues were taken from a 7,12-dimethylbenz[a]anthracene-induced hamster buccal pouch carcinogenesis model. Analyses were conducted at various excitation wavelengths, ranging from 280 nm to 400 nm in 20 nm increments, to assess classification performance at different excitations. For each excitation the PLS analysis and logistic regression were combined, on the basis of cross validation, to calculate the posterior probabilities of samples belonging to four stages of cancer development: normal tissues, hyperplasia, dysplasia and early cancers and frankly invasive cancers. Results showed that the 320 nm excitation wavelength optimally classified the cancer development stages: the accuracy rates for identifying samples at that excitation were 91.7%, 83.3%, 66.7% and 83.3% for the four respective stages. The average accuracy rate was 81.3%. These results suggest that the algorithm described in this study might be useful for the detection of human oral cancers.     

应用遗传算法和PLS的近红外光谱预测玉米中淀粉含量的研究

以普通玉米籽粒为试验材料,在应用遗传算法结合偏最小二乘回归法对近红外光谱数据进行特征波长选择的基础上,应用偏最小二乘回归法建立了特征波长测定玉米籽粒中淀粉含量的校正模型．试验结果表明,基于11个特征波长所建立的校正模型,其校正误差（RMSEC）、交叉检验误差（RMSECV）和预测误差（RMSEP）分别为0.30％、0.35％和0.27％,校正数据集和独立的检验数据集的预测值与实际测定值之间的相关系数分别达到0.9279和0.9390,与全光谱数据所建立的预测模型相比,在预测精度上均有所改善,表明应用遗传算法和PLS进行光谱特征选择,能获得更简单和更好的模型,为玉米籽粒中淀粉含量的近红外测定和红外光谱数据的处理提供了新的方法与途径．