基于可见-近红外光谱识别氧乐果污染的脐橙

该文对喷施过不同浓度氧乐果农药的脐橙样品采集可见-近红外光谱进行识别。采用多元散射校正（MSC）、标准正交变量变换（SNV）、一阶导数（FD）和二阶导数（SD）4种不同光谱预处理方法时,分别选取430～1 000、1 000～1 800和430～1 800 nm 3种波谱范围建立偏最小二乘法（PLS）农药污染预测模型。比较实验结果表明：波谱范围取430～1 000 nm,采用一阶导数的预处理方法时应用建立的PLS预测模型最优,其验证组脐橙表面氧乐果污染程度的实际类别与预测类别的相关系数Rpred为0.9817,预测样本均方根误差RMSEP是0.1564。     

近红外光谱快速检测食用油必需脂肪酸

为了建立食用油必需脂肪酸快速检测的方法,该研究提出了基于近红外光谱技术检测食用油中α-亚麻酸和亚油酸含量的快速测定方法。对光谱信息分别采用偏最小二乘回归方法(PLS)和最小二乘支持向量机(LS-SVM)建立模型。比较了多种光谱预处理方法对模型预测能力的影响。结果表明对于亚油酸含量的预测,采用Savitzky-Golay平滑法结合多元散射校正(MSC)的光谱预处理所建立的LS-SVM模型最优。预测集的决定系数(R2)、预测均方根误差(RMSEP)和剩余预测偏差(RPD)分别达到了0.989,0.0161和9.4783。对于α-亚麻酸含量的预测,采用Savitzky-Golay平滑法结合标准正态变换(SNV)的光谱预处理所建立的LS-SVM模型最优。α-亚麻酸含量预测结果的R2、RMSEP和RPD为0.972,0.0036和6.0561,据此表明,应用近红外光谱技术能够检测食用油中α-亚麻酸和亚油酸的含量,为快速检测食用油的必需脂肪酸提供了参考。     

除草剂胁迫下大麦叶片丙二醛含量的光谱快速检测方法

丙二醛(MDA)是植物衰老和抗性生理研究中的一个重要指标,传统检测方法程序复杂,检测费时。该研究应用近红外光谱技术实现了除草剂胁迫下大麦叶片丙二醛(MDA)含量的简便、无损、快速检测。采集75个大麦叶片样本的近红外光谱数据,比较了Savitzky-Golay平滑(SG)、变量标准化(SNV)、多元散射校正(MSC)等7种预处理方法,建立了大麦叶片丙二醛含量预测的最优偏最小二乘法(PLS)模型,将PLS提取的特征向量(LV)作为最小二乘-支持向量机(LS-SVM)模型的输入变量,建立了LV-LS-SVM模型。选用回归系数(RC)方法提取原始光谱的特征波长,将其分别作为PLS、MLR和LS-SVM的输入变量建立相应模型。将相关系数(r)和预测集均方根误差(RMSEP)作为模型的主要评价指标。结果表明,LV-LS-SVM模型效果优于PLS模型,LV-LS-SVM模型在SNV及MSC预处理后预测效果相同,其预测的r和RMSEP分别为0.9383和10.4598,获得了满意的预测效果。说明应用光谱技术检测大麦叶片中MDA含量是可行的,且预测精度较高,为大麦生长状况的大田监测及除草剂胁迫对大麦抗性等生理信息的快速检测提供了新的途径。     

Use of visible and near infrared spectroscopy and least squares-support vector machine to determine soluble solids content and pH of cola beverage

Visible and near infrared (VIS/NIR) transmission spectroscopy and chemometric methods were utilized for the fast determination of soluble solids content (SSC) and pH of cola beverage. A total of 180 samples were used for the calibration set, whereas 60 samples were used for the validation set. Some preprocessing methods were applied before developing the calibration models. Several PLS factors, extracted by partial least squares (PLS) analysis, were used as the inputs of least squares-support vector machine (LS-SVM) model according to their accumulative reliabilities. The correlation coefficient (r), root mean square error of prediction (rmsEP), bias, and RPD were 0.959, 1.136, -0.185, and 3.5 for SSC, whereas 0.973, 0.053, 0.017, and 4.1 for pH, respectively. An excellent prediction precision was achieved by LS-SVM compared with PLS. The results indicated that VIS/NIR spectroscopy combined with LS-SVM could be applied as a rapid and alternative way for the fast determination of SSC and pH of cola beverage.     

表面增强拉曼光谱检测脐橙果皮混合农药残留

为了研究果皮农药残留快速检测方法。该文以脐橙为例,混合农药(亚胺硫磷和乐果)为研究对象,选用银纳米线作为增强基底,利用共焦显微拉曼光谱仪对农药残留进行检测。通过表面增强拉曼光谱(surface enhanced Raman scattering,SERS)技术,采集脐橙表皮混合农药残留的SERS光谱。对混合农药定性分析,银纳米线对2种农药都有较好的增强效果。对采集的光谱进行预处理后,建立模型,进行定量分析,研究结果表明,经过二阶微分预处理后光谱数据结合偏最小二乘法(partial least squares,PLS)得到的模型预测效果最好,预测相关系数(R_p)为0.954,其预测均方根误差(root-mean-square prediction error,RMSEP)为4.822 mg/L。挑选两种农药特征峰的特征波段,混合农药中亚胺硫磷的特征波段经多元散射校正(multiplicative scatter correction,MSC)处理后,建模效果较好,其中R_p为0.898,RMSEP为6.621 mg/L;混合农药中乐果的特征波段经基线校正处理后,建模效果较好,其中R_p为0.911,RMSEP为7.369 mg/L。研究结果表明SERS技术是一种快速、可靠的检测混合农药残留的方法。     

便携式双档位柑橘多品质无损检测装置设计与试验

为了解决普通检测装置难以覆盖不同果径（25～95mm）柑橘的检测需求问题,研发了覆盖多果径柑橘的便携式双档位多品质无损检测装置。以砂糖橘（果径25.35～48.61mm）和武鸣沃柑（果径53.24～94.71mm）为研究对象,基于研发的双档位探头,在赤道部位每隔120°采集一次光谱,平均光谱作为该柑橘的原始光谱。经标准正态变量变换（standard normal variable,SNV）、多元散射校正（multiplicative scatter correction,MSC）预处理,再利用竞争性自适应加权抽样算法（competitive adaptive reweighted sampling, CARS）筛选特征波长,分别建立了沃柑和砂糖橘的可溶性固形物含量（soluble solids content,SSC）和水分的偏最小二乘预测模型。沃柑的SSC和水分预测模型验证集相关系数分别为0.937、0.951,均方根误差分别为0.382°Brix、0.491%;砂糖橘的SSC和水分预测模型验证集相关系数分别为0.921、0.935,均方根误差分别为0.460°Brix、0.673%...     

Chemometric characterization of fruit juices from Spanish cultivars according to their phenolic compound contents: I. Citrus fruits

The data set composed by phenolic compound profiles of 83 Citrus juices (determined by HPLC-DAD-MS/MS) was evaluated by chemometrics to differentiate them according to Citrus species (sweet orange, tangerine, lemon, and grapefruit). Cluster analysis (CA) and principal component analysis (PCA) showed natural sample grouping among Citrus species and even the Citrus subclass. Most of the information contained in the full data set can be captured if only 15 phenolic compounds (concentration ≥10 mg/L), which can be quantified with fast and accurate methods in real samples, are introduced in the models; a good classification which allows the confirmation of the authenticity of juices is achieved by linear discriminant analysis. Using this reduced data set, fast and routine methods have been developed for predicting the percentage of grapefruit in adulterated sweet orange juices using principal component regression (PCR) and partial least-squares regression (PLS). The PLS model has provided suitable estimation errors.     

基于小波变换的番茄总糖近红外无损检测

分别采用小波消噪、常数偏移消除等11种光谱预处理方法,对番茄总糖含量（质量分数）的近红外光谱进行预处理,通过偏最小二乘法定量校正模型预测值比较得出,小波消噪是适合番茄近红外光谱的最佳预处理方法,小波消噪的总糖质量分数近红外光谱优选区域为11 998.9～6 097.8 cm-1和4 601.3～4 246.5 cm-1,在此光谱区内建立的番茄总糖质量分数偏最小二乘法模型预测值与实测值的相关系数为0.930,内部交叉验证均方差为0.466%,校正标准差为0.469%,预测标准差为0.260%。试验结果表明：小波消噪后建立的近红外光谱模型能准确地对番茄总糖含量进行快速无损检测。     

基于近红外光谱技术的淡水鱼品种快速鉴别

为探索淡水鱼品种的快速鉴别方法,该文应用近红外光谱分析技术,结合化学计量学方法,对7种淡水鱼品种的判别分类进行了研究。采集了青、草、鲢、鳙、鲤、鲫、鲂等7种淡水鱼,共665个鱼肉样品的近红外光谱数据,经过多元散射校正(multiplicative scatter correction,MSC)、正交信号校正(orthogonal signal correction,OSC)、数据标准化(standardization,S)等20种方法预处理,在1 000~1 799 nm范围内分别采用偏最小二乘法(partial least square,PLS)、主成分分析(principal component analysis,PCA)和BP人工神经网络技术(back propagation artificial neural network,BP-ANN)、偏最小二乘法和BP人工神经网络技术对7种淡水鱼原始光谱数据进行了鉴别分析。结果表明,近红外光谱数据,结合主成分分析和BP人工神经网络技术建立的淡水鱼品种鉴别模型最优,模型的鉴别准确率达96.4%,对未知样本的鉴别准确率达95.5%。模型具有较好的鉴别能力,采用该方法能较为准确、快速地鉴别出淡水鱼的品种。     

基于RGB图像光谱重建的鱼糜掺假定量检测研究

实现基于RGB图像的光谱重建对降低光谱的硬件要求、扩大其实际应用具有重大意义。该研究以鱼糜掺假检测为例,比较多元多项式最小二乘回归算法（polynomial multivariate least-squares regression,PMLR）与深度学习HRNet网络对光谱重建的性能,建立基于重建光谱多种掺假鱼糜检测模型并验证其实际应用的有效性。结果表明,2种方法的重建光谱误差较小,HRNet网络、PMLR算法重建光谱的均方根误差（root mean square error,RMSE）分别为0.010 4和0.012 6,大多数掺假检测模型有较高的预测准确性,其预测相关系数大于0.91,预测均方根误差小于9%。在基于重建光谱建立的掺假检测模型中,效果最佳的是基于PMLR算法重建光谱使用标准正态变量变换（standard normal variate,SNV）预处理的极限学习机回归模型,其预测均方根误差为3.954 4%、预测相关系数为0.983 0。因此,PMLR算法和HRNet网络均能较好的实现基于RGB图像的光谱重建,且重建光谱均能实现对鱼糜掺假样本的较好检测结果,为基于重建光谱的食品和农产品品质与安全检测提供了新思路。     

基于高光谱成像技术的青贮玉米饲料pH值无损检测

为实现青贮玉米饲料pH值的快速、无损检测,该研究采用高光谱成像技术建立不同品质青贮玉米饲料pH值的定量检测模型。采集青贮玉米饲料样本936～2 539nm的平均光谱,采用6种预处理方法对青贮玉米饲料平均光谱进行处理,通过建立偏最小二乘回归（partialleastsquaresregression,PLSR）模型得出多元散射校正（multiplicativescatter correction,MSC）和卷积平滑（savitzky-golay,SG）两种预处理方法效果较好,使用竞争性自适应重加权算法（competitive adaptive reweighted sampling,CARS）、变量组合集群分析算法（variable combination population analysis,VCPA）以及迭代保留信息变量（iteratively retains informative variables,IRIV）算法对经MSC和SG卷积平滑预处理光谱进行特征波长提取,利用PLSR和极限学习机（extreme learning machines,ELM）分别建立饲料全波段、特征波长...     

Predicting the age and type of tuocha tea by fourier transform infrared spectroscopy and chemometric data analysis

Fourier transform infrared (FTIR) spectroscopy combined with chemometric multivariate methods was proposed to discriminate the type (unfermented and fermented) and predict the age of tuocha tea. Transmittance FTIR spectra ranging from 400 to 4000 cm(-1) of 80 fermented and 98 unfermented tea samples from Yunnan province of China were measured. Sample preparation involved finely grinding tea samples and formation of thin KBr disks (under 120 kg/cm(2) for 5 min). For data analysis, partial least-squares (PLS) discriminant analysis (PLSDA) was applied to discriminate unfermented and fermented teas. The sensitivity and specificity of PLSDA with first-derivative spectra were 93 and 96%, respectively. Multivariate calibration models were developed to predict the age of fermented and unfermented teas. Different options of data preprocessing and calibration models were investigated. Whereas linear PLS based on standard normal variate (SNV) spectra was adequate for modeling the age of unfermented tea samples (RMSEP = 1.47 months), a nonlinear back-propagation-artificial neutral network was required for calibrating the age of fermented tea (RMSEP = 1.67 months with second-derivative spectra). For type discrimination and calibration of tea age, SNV and derivative preprocessing played an important role in reducing the spectral variations caused by scattering effects and baseline shifts.     

Quantitation of curcuminoids in curcuma rhizome by near-infrared spectroscopic analysis

This study investigated a nondestructive and rapid quantitation method for the curcuminoids, including curcumin, demethoxycurcumin, and bisdemethoxycurcumin, present in turmeric using near-infrared (NIR) spectroscopy and multivariate statistics. In the second derivatives of the NIR spectra of turmeric samples, two characteristic absorptions of curcuminoids were detected around 1700 and 2300-2320 nm. Partial least-squares regression (PLS-R) analysis was applied to the NIR spectra obtained from 34 turmeric samples, and PLS models for the quantitation of curcumin, demethoxycurcumin, bisdemethoxycurcumin, and total curcuminoid contents in the pulverized turmeric samples were constructed. Combination usage of the Standard Normal Variate (SNV) and second derivatives was obviously superior to other preprocessing methods. The lowest root mean squared error of cross-validation (RMSECV) values were detected at 6, 6, 6, and 6 PLS factors, for the quantitative subjects curcumin, demethoxycurcumin, bisdemethoxycurcumin, and total curcuminoid contents. It was clarified that the prediction of the composition by PLS-R analysis showed high correlation with the results of HPLC quantitations.     

基于冠层光谱的锦橙叶片磷素营养监测研究

以盆栽蓬安100号锦橙施肥调控试验为基础,利用田间冠层光谱信息探索建立植株磷素营养监测技术与方法。通过采集蓬安100号锦橙95个单株样本的冠层光谱信息和室内检测分析叶片磷含量,随机选取76个作为建模样本,19个为检验样本,运用多种光谱预处理方法和偏最小二乘法(Partial least square method,PLS)及内部交叉验证方法建立校正模型与模型检验。结果表明,经多种光谱预处理方法的建模结果比较,冠层原始反射光谱经二阶求导和SNV处理后建立的蓬安100号锦橙叶片磷含量冠层光谱监测模型预测能力和稳健性最佳,其主成分数4个,能表达全波段63%的信息;校正模型相关系数为0.90,偏差Bias=2.45E-10,且RMSEC和RMSEP均最小。模型检验预测的决定系数R2=0.85。因此,利用二阶导数及标准归一化(Standard normal variate transformation,SNV)预处理的田间冠层光谱信息快速无损监测蓬安100号锦橙叶片磷含量具有一定的可行性。     

北京典型耕作土壤养分的近红外光谱分析

为研究土壤养分含量分布信息,以从北京郊区一块试验田采集的72个土壤样品为试验材料,应用傅里叶变换近红外光谱技术分析了土样的全氮、全钾、有机质养分含量和pH值。采用偏最小二乘法(PLS)对光谱数据与土壤养分实测值进行回归分析,建立预测模型,以模型决定系数(R2)、校正标准差(RMSECV)、预测标准差(RMSEP)和相对分析误差(RPD)作为模型精度的评价指标。结果表明,利用该模型与光谱数据对土壤全氮、全钾、有机质养分含量和pH值进行预测,结果与实测数据具有较好的一致性,最高决定系数R2达到0.9544。偏最小二乘回归方法建立的养分预测模型能准确地对北京地区褐土土质全氮、有机质、全钾和pH值4种养分进行预测。     

基于可见/近红外反射光谱的梨树叶片钾含量的快速测定研究

利用可见/近红外反射光谱定量分析技术对梨树鲜叶钾素含量进行快速测定研究。对150个梨树叶片样本进行光谱扫描,其中120个做建模集,30个做验证集。通过对样品的可见/近红外光谱进行多种预处理,并建立钾素预测模型,探讨了可见/近红外光谱数据预处理对预测精度的影响。结果表明,通过原始光谱与S-G(3)平滑相结合的预处理方法,用17个主成分建立的偏最小二乘法模型最好,其交叉验证集和预测集模型的决定系数(R2)分别为0.722 7和0.679 1,交叉验证均方根误差(RMSECV)为1.171,预测的平均相对误差为6.81%,能高效、快速地预测梨树叶片钾素含量,为梨树钾素快速测定提供了新的手段。     

Comparison of the HPLC method and FT-NIR analysis for quantification of glucose, fructose, and sucrose in intact apple fruits

A rapid quantification method was developed and validated for simultaneous and nondestructive quantifying the constituent sugar concentrations of intact apples using Fourier transform near-infrared (FT-NIR) spectroscopy in diffuse reflectance mode. Multiplicative scatter correction (MSC), the second derivative of Savitsky-Golay, and mean centering were used as spectral preprocessing options. Calibration models were established by the partial least squares (PLS) regression analysis, and validation of the method was performed according to the high-performance liquid chromatography (HPLC) chromatographic method. Spectral range and the number of PLS factors were optimized for the lowest root-mean-square error of prediction (RMSEP) and correlation coefficient of determination (r). The best models showed satisfactory predictions as measured by the RMSEP and r values: glucose, 0.201 and 0.950; fructose, 0.298 and 0.968; sucrose, 0.335 and 0.969, respectively. FT-NIR analysis of constituent sugar concentrations in the intact apple form was found to be more flexible and much faster than performed with the HPLC method.     

基于1D-CNN的土壤全氮近红外光谱预测模型

摘要：【目的】传统的基于近红外光谱数据预测土壤全氮的方法需要对原始光谱数据做复杂的预处理，筛选出与土壤全氮含量相关性高的敏感波长之后进行模型的回归拟合。本文提出一种一维卷积神经网络（1D-CNN）模型，可以在对数据进行简单预处理甚至无处理的情况下达到非常理想的结果，实现用近红外光谱技术对土壤全氮含量的预测。【方法】于江苏无锡采集410个土壤样品，利用半微量开氏法(NY/T 53-1987)测定土壤的全氮含量，并利用NIR Quest 512光谱仪，在室内环境下对每份土壤样品做光谱检测，并用均值中心化(CT)、标准正态变换(SNV)、趋势校正(DT)对光谱进行预处理，运用偏最小二乘回归（PLS）、BP神经网络、1D-CNN方法建立土壤全氮含量的回归预测模型。每种模型在采用不同预处理方法的数据集上做十折交叉验证，记录预测模型的决定系数(R2)和均方根误差(RMSE)的平均值，并对比三种预处理方法对模型精度的影响。【结果】证明了本文提出的1D-CNN模型基于土壤近红外光谱数据预测土壤全氮含量的可靠性。使用原始数据与经均值中心化、标准正态变换、趋势校正预处理的数据训练得到的1D-CNN模型的决定系数分别为0.907、0.931、0.922、0.964，构建的PLS回归模型决定系数为0.856、0.863、0.861、0.880，训练的BP神经网络的决定系数为0.874、0.907、0.901、0.911。【结论】本文提出的1D-CNN模型在原始数据和经预处理的光谱数据上的表现都优于PLS和BP神经网络，且可以证明，对光谱数据进行预处理能够有效提高1D-CNN模型的性能，尤其是趋势校正对模型的提升效果最明显。研究表明，1D-CNN能更好地提取光谱特征并建立其与含氮量的映射关系，有效地避免过拟合，在未经过预处理的光谱数据上依然能够达到一定的精度。     

鸭肉中谷氨酸含量的可见-近红外光谱测定研究

为实现鸭肉中谷氨酸含量的快速测定,提出了利用可见-近红外光谱结合PCA、BP神经网络来建立鸭肉中谷氨酸含量测定的定量分析模型.采集试验首先采集光谱范围在350～1800nm的鸭肉可见-近红外反射光谱,并在430～1000、1001～1400和430～1400nm 3个光谱范围内分别用一阶导数(FD)、二阶导数(SD)、...     

Estimating hydrogen cyanide in forage sorghum ( Sorghum bicolor ) by near-infrared spectroscopy

Hydrogen cyanide (HCN) is a toxic chemical that can potentially cause mild to severe reactions in animals when grazing forage sorghum. Developing technologies to monitor the level of HCN in the growing crop would benefit graziers, so that they can move cattle into paddocks with acceptable levels of HCN. In this study, we developed near-infrared spectroscopy (NIRS) calibrations to estimate HCN in forage sorghum and hay. The full spectral NIRS range (400-2498 nm) was used as well as specific spectral ranges within the full spectral range, i.e., visible (400-750 nm), shortwave (800-1100 nm) and near-infrared (NIR) (1100-2498 nm). Using the full spectrum approach and partial least-squares (PLS), the calibration produced a coefficient of determination (R(2)) = 0.838 and standard error of cross-validation (SECV) = 0.040%, while the validation set had a R(2) = 0.824 with a low standard error of prediction (SEP = 0.047%). When using a multiple linear regression (MLR) approach, the best model (NIR spectra) produced a R(2) = 0.847 and standard error of calibration (SEC) = 0.050% and a R(2) = 0.829 and SEP = 0.057% for the validation set. The MLR models built from these spectral regions all used nine wavelengths. Two specific wavelengths 2034 and 2458 nm were of interest, with the former associated with C═O carbonyl stretch and the latter associated with C-N-C stretching. The most accurate PLS and MLR models produced a ratio of standard error of prediction to standard deviation of 3.4 and 3.0, respectively, suggesting that the calibrations could be used for screening breeding material. The results indicated that it should be feasible to develop calibrations using PLS or MLR models for a number of users, including breeding programs to screen for genotypes with low HCN, as well as graziers to monitor crop status to help with grazing efficiency.