食用向日葵子仁蛋白质含量近红外光谱模型的建立

以自育的57份食用向日葵子仁为测试对象,用化学方法测定蛋白质含量,对照近红外漫反射光谱,通过偏最小二乘法建立了食用向日葵子仁蛋白质含量的近红外光谱模型。结果表明,对原始光谱数据采用“一阶倒数+多元散射校正(FD+MSC)”处理的方法建立的模型其校正或预侧效果最佳。该模型的校正决定系数和验证决定系数分别为0.95和0.93,校正标准误和预测标准误分别为0.96和1.16。用该模型对16份未参与建模的食用向日葵材料进行了预测,结果表明该模型预测能力较好。     

基于高光谱成像技术快速无损测定花生中水分含量

花生中水分含量的高低直接影响花生及其制品的贮藏期,而现有的测定方法存在步骤多、时间长等问题。试验利用高光谱成像技术对花生中水分含量进行快速无损检测分析。通过采集120个花生样品的图像信息,从校正后的图像中提取花生目标区域的平均光谱作为花生光谱信息进行分析;同时,优选最佳的光谱预处理方法和建模方法建立花生中水分含量全波段模型,在此基础上利用回归系数法,确定重要波长并建立模型。结果表明,二阶导数(2nd-der)偏最小二乘法(PLS)全波段模型预测水分含量能力最佳,校正集和预测集的相关系数分别为0.91和0.84,标准偏差分别为0.28和0.38;回归系数法确定的14个波长所建简化模型的性能与全波段相当,校正集和预测集的相关系数分别为0.82和0.81,标准偏差分别为0.39和0.43。因此,高光谱成像技术可以快速无损测定花生中水分含量,其具有快速运算特点的重要波长模型可以更加方便地应用于花生加工产业中。     

基于 SPA-RBF神经网络的小麦蛋白质含量无损检测

传统半微量凯氏法测量小麦蛋白质含量繁琐费时,应用近红外光谱分析技术结合SPA-RBF神经网络对小麦蛋白质含量进行快速、无损检测.采用SPXY算法划分校正集和预测集样本,运用连续投影算法(SPA)对一阶微分和SNV预处理后的光谱数据提取敏感波点作为RBF神经网络的输入,建立小麦蛋白质含量的SPA-RBF神经网络校正模型.模型的预测均方根误差和预测相关系数可达到0.26576和0.975,预测效果较好,基本上可以完成粮食储备和食品加工行业对小麦及其制品品质的划分以及育种上的前期世代筛选.研究表明:近红外光谱技术结合SPA-RBF神经网络可实现对小麦蛋白质含量的检测,满足现代农业发展对小麦无损、实时、大量检测的需要.     

基于近红外光谱的棉花毛籽蛋白质和油分含量快速检测

【目的】建立棉花毛籽蛋白质和油分含量的近红外检测校正模型。【方法】检测样本的蛋白质含量和油分含量，根据光谱-理化值共生距离算法（sample set partitioning based on joint X-Y distance sampling, SPXY）按照3∶1的比例将426个样本划分为包含320个样本的校正集和106个样本的预测集，结合多元散射校正和一阶导数等光谱预处理方法对模型进行优化，并采用线性偏最小二乘法（partial least square method, PLS）、支持向量机（support vector machine, SVM）和随机森林（random forest, RF）3种方法对比分析建立棉花毛籽蛋白质和油分含量的近红外快速测定模型，以决定系数、均方根误差和剩余预测偏差作为模型的评价指标。【结果】SVM模型和PLS模型在校正集的拟合效果较好，决定系数均大于0.8，但对预测集的拟合决定系数不到0.8，说明模型均存在过拟合现象；而RF模型在校正集和预测集的拟合效果都非常好，决定系数均大于0.9，其中蛋白质含量预测模型的决定系数、预测均方根误差和剩余预测偏...     

基于连续投影算法的小麦湿面筋近红外校正模型优化

为减少建模过程中的计算量、提高模型的稳健性及预测精度,将连续投影算法用于小麦湿面筋近红外校正模型的建立。首先采用SPXY算法选择具有代表性的校正集样本,然后对光谱数据作不同预处理,增强光谱特征;运用连续投影算法对原始光谱和预处理后的光谱进行敏感波点提取,进而分别建立多元线性回归校正模型。测试结果表明,对光谱标准正态变量变换后利用连续投影算法提取敏感波点所建多元线性回归模型预测效果最好,预测均方根误差和预测相关系数分别为1.3332和0.94319,优于同等条件下建立的偏最小二乘回归模型。     

基于近红外透射光谱分析技术的小麦蛋白质含量测定

近红外分析技术具有快速、简便、准确、非破坏性的优点,为小麦品质检测提供了一个新的技术手段。笔者首先对光谱数据进行移动窗口平均平滑和马氏距离筛选,在以上光谱预处理基础上,建立了小麦籽粒蛋白的偏最小二乘校正模型。仿真试验结果表明：该模型能够较准确地预测小麦蛋白质的含量,预测相关系数、预测均方误差和平均相对误差分别为0.9809、0.1130、1.973%。与用原始数据所建校正模型相比,预测效果显著提高。     

基于遗传算法的BP神经网络在小麦蛋白质含量预测中的应用

为了快速、简便、准确地测定小麦蛋白质的含量,本文提出了应用近红外光谱分析技术结合遗传算法(GA)的BP神经网络的建模方法。采用SPXY算法对光谱数据进行了合理划分,并运用连续投影算法(SPA)将预处理过的数据压缩,对光谱数据提取最佳敏感波点作为GA-BP神经网络的输入,建立小麦蛋白质含量的校正模型。模型的预测均方根误差和预测相关系数为1.3379和0.979,并与BP神经网络所建立的校正模型进行了比较。结果表明：GA-BP神经网络所建模型收敛速度快、训练时间短、准确度也较高,能够实现对小麦蛋白质含量快速高效的检测。     

利用VIS/NIR反射光谱快速检测柿饼加工中水分含量

水分是柿饼的重要组成成分,也是影响柿饼制作过程的重要因素。利用可见/近红外反射光谱对柿饼制作过程中的水分含量进行检测。首先,获取柿饼在不同加工阶段的可见/近红外反射光谱(400～1 000 nm),采用烘干法测定柿饼水分含量。然后,对光谱进行Mean smoothing (MS)平滑、多元散射校正(MSC)和一阶导数(1-D)预处理。最后,对不同预处理光谱,结合样本水分含量,使用Samples set partitioning based on joint x-y distance (SPXY)方法划分校正集和验证集,基于SPA方法选择特征波长,建立多元线性回归(MLR)预测模型。结果表明,反射光谱经过MS处理后,确定的9个最优波长组合建立水分检测模型的预测结果最好:预测相关系数(Rp)为0.969 0,预测标准残差(SEP)为3.472 9%,可见/近红外反射光谱技术可以较好地预测柿饼制作过程中的的水分含量。研究可为柿饼加工过程中的品质快速检测提供一定的技术支撑。     

近红外光谱无损检测石榴中花色苷含量研究

以全国不同地区的97个石榴为样本,研究近红外光谱无损检测石榴中花色苷的含量,探讨了不同数据处理和回归方法对建模效果的影响。结果表明,对原始光谱进行一阶微分、标准多元离散校正法处理后,采用偏最小二乘法建立的石榴花色苷含量预测模型,预测偏差为0.148,预测标准差(SEP)为1.47,相关系数为0.829,模型预测良好,说明近红外光谱无损检测石榴的品质是可行的。     

应用近红外光谱技术分析稻米蛋白质含量

以稻谷、米粒、米粉3种形态的样品，应用近红外光谱技术(NIRS)和偏最小二阶乘法(PLS)，建立了6个稻米蛋白质含量近红外光谱数学模型，并对模型预测结果的准确性进行了评价。结果表明，糙米蛋白质含量的稻谷、糙米粒和糙米粉近红外光谱预测模型校正决定系数(R_C²)分别为0.893、0.971和0.987，校正标准差(RMSEC)分别为0.507、0.259和0.183；精米蛋白质含量的稻谷、精米粒和精米粉近红外光谱预测模型R_C²分别为0.897、0.984和0.986，RMSEC分别为0.497、0.186和0.190。模型内部交叉验证分析表明，预测糙米蛋白含量的稻谷、糙米粒和糙米粉模型内部交叉验证决定系数(R_CV²)分别为0.865、0.962和0.984，内部验证标准差(RMSECV)分别为0.557、0.290和0.205；预测精米蛋白含量的稻谷、精米粒和精米粉的模型R_CV²分别为0.845、0.951和0.979，RMSECV分别为0.594、0.316和0.233。模型外部验证分析表明，预测糙米蛋白含量的稻谷、糙米粒和糙米粉近红外光谱模型外部验证决定系数(R_V²)分别为0.683、0.801和0.939，外部验证标准差(RMSEV)为0.962、0.799和0.434；预测精米蛋白含量的稻谷、精米粒和精米粉近红外光谱的模型R_V²分别为0.673、0.921和0.959，RMSEV为0.976、0.513和0.344。用米粉建立的近红外光谱预模型准确性最高，米粒次之，基于稻谷的预测模型准确性相对较低；内部交叉验证和外部验证表明，近红外光谱分析技术与化学分析方法一致性较好，且能保证样品的完整性，在水稻优质育种和稻米品质分析中具有广泛的应用价值。     

近红外光谱法测定大米中的淀粉含量

用化学方法测定64个大米样品中的淀粉含量,利用近红外谷物分析仪采集样品的近红外光谱,选择合适的光谱区间和光谱预处理方法。50个定标集样品的近红外光谱经二阶导数及标准多元离散校正(Standard MSC)预处理,结合偏最小二乘法(PLS)建立了大米中的淀粉含量测定的定标模型,其相关系数为0.8780。14个验证集样品用于外部检验,大米中的淀粉含量的模型预测值与化学值之间的相关系数为0.9498。     

近红外光谱法非破坏性测定玉米子粒粗淀粉含量的研究

采用偏最小二乘回归法,对近红外光谱法测定玉米完整子粒粗淀粉含量的可行性进行研究。结果表明,定标集和检验集的预测值与化学测定值间均达极显著正相关,相关系数分别为0.9610和0.9820,并具有较小的定标标准差和预测标准差,分别为0.707和0.666。所建立的校正模型具有较高的预测精度。     

快速溶剂萃取-气相色谱-质谱法测定土壤中15种硝基苯类化合物

通过气相色谱-质谱法(GC-MS),建立土壤中15 种硝基苯类化合物的快速测定方法,以期为土壤和固体物中硝基苯类化合物残留量的测定提供参考。样品中的15 种硝基苯类化合物经正己烷和丙酮(1:1, V:V)溶剂萃取,硅酸镁小柱净化,10 mL正己烷:丙酮混合溶液(1:1, V:V)洗脱并干燥浓缩后,用GCMS测定,内标法定量。结果表明：此方法线性关系良好,相关系数均大于0.995,15 种硝基苯类化合物的回收率在71.6%~119%之间,相对标准偏差在0.2%~1.1%之间,检出限在0.05~1.31 μg/kg 之间。该方法简便、准确度好、灵敏度高,节约人力财力,满足土壤中硝基苯类化合物的测定要求。     

Estimation of seed weight, oil content and fatty acid composition in intact single seeds of rapeseed ( Brassica napus} L.) by near-infrared reflectance spectroscopy

The potential of near-infrared reflectance spectroscopy (NIRS) for the simultaneous analysis of seed weight, total oil content and its fatty acid composition in intact single seeds of rapeseed was studied. A calibration set of 530 single seeds was analysed by both NIRS and gas-liquid chromatography (GLC) and calibration equations for the major fatty acids were developed. External validation with a set of 75 seeds demonstrated a close relationship between NIRS and GLC data for oleic (r = 0.92) and erucic acid (r = 0.94), but not for linoleic (r = 0.75) and linolenic acid (r = 0.73). Calibration equations for seed weight and oil content were developed from a calibration set of 125 seeds. A gravimetric determination was used as reference method for oil content. External validation revealed a coefficient of correlation between NIRS and reference methods of 0.92 for both traits. The performance of the calibration equations for oleic and erucic acid was further studied by analysing two segregating F2 seed populations not represented in the calibration set. The results demonstrated that a reliable selection for both fatty acids in segregating populations can be made by using NIRS. We concluded that a reliable estimation of seed weight, oil content, oleic acid and erucic acid content in intact, single seeds of rapeseed is possible by using NIRS technique. This revised version was published online in July 2006 with corrections to the Cover Date.     

Analysis of total glucosinolate content and individual glucosinolates in Brassica spp. by near-infrared reflectance spectroscopy

This study was conducted to test the applicability of near-infrared reflectance spectroscopy (NIRS) for estimating the total glucosinolate (GSL) content in samples of intact seed from a wide range of Brassica species, and to develop calibration equations to estimate simultaneously the percentage of individual GSLs. A total of 290 samples from 15 different Brassica species were scanned by NIRS and analysed for glucosinolate content by high-pressure liquid chromatography (HPLC). A calibration equation for total GSL content was developed using 270 samples of 14 species in a range between 6 and 193μmol/g seed, resulting in an r² of 0.99 in calibration and cross-validation, and 0.95 in independent validation with 20 samples of Brassica rapa, a species not represented in the calibration. Furthermore, calibration equations to estimate the relative amount (mol/mol) of progoitrin, sinigrin, and gluconapin were successfully developed (r² > 0.85 in cross-validation) and validated with samples from species not included in the calibration. It was also possible to discriminate between entries with high and low values of glucoiberin, 4-hydroxyglucobrassicin and glucoerucin.     

毛细管气相色谱内标法测定防霉剂中富马酸二甲酯

建立了巴豆酸为内标物,用毛细管气相色谱快速测定防霉剂中富马酸二甲酯含量的方法.样品用乙酸乙酯提取,以DM-FFAP毛细管柱分离,火焰离子化检测器检测.结果表明,在0.25～3.0 mg/mL浓度范围内,富马酸二甲酯和巴豆酸的浓度比与峰面积比的回归方程为y=-0.0009 0.7174x,相关系数r=0.9996,检出限为5.0 mg/L;样品加标回收率(n=6)为97.0%～104.0%,相对标准偏差为3.26%.该方法简单、快速、准确度好,可用于测定防霉剂中的富马酸二甲酯含量.     

基于主成分回归的苹果可溶性固形物含量预测模型

为了更好地预测苹果的可溶性固形物含量（SSC）,试验采用反射式光谱采集系统获取采后“富士”苹果的光谱反射率。分析了3种光谱预处理方法（标准正态变换、多元散射校正以及二阶导数）对预测模型的影响;利用主成分分析方法对预处理后的光谱数据进行降维,并基于选取的特征变量建立预测苹果SSC的回归模型。结果表明：采用主成分分析方法从全光谱的1 024个波长中选取了前23个主成分得分作为特征变量;基于特征变量建立的回归预测模型具有较好的预测能力,其预测集相关系数RP=0.908,均方根误差RMSEP=0.499。这表明采用光谱技术结合主成分回归预测苹果SSC是可行的。     

近红外（NIR）光谱法测定完整苹果糖的含量

介绍一种快速分析完整苹果糖分的新方法──近红外光谱法。在波长９１０ｎｍ附近，高、中、低糖含量的二阶导数光谱之间有明显差异，该波长选作定标的第一波长。经９１０ｎｍ、８８４ｎｍ、８４３ｎｍ、９９１ｎｍ四波长线性回归分析，其相关系数为０．９８４，标准误差为０．３６０，检验时的标准误差为０．４５０，离差为０．１１。ＮＩＲ光谱法在实际应用中可满足完整苹果糖含量的测定精度。     

Near infrared reflectance spectroscopy analysis of digestibility traits of maize stover

Summary Near infrared reflectance spectroscopy (NIRS) is emerging as a potentially useful tool in breeding plants for quality traits. Information is lacking, however, on its use in forage maize (Zea mays L.). The objectives of the present investigation were to evaluate the prediction of digestibility traits of maize stover using NIRS technique and to study the effect of laboratory (Lab) and NIRS assays on the estimates of variation and covariation. Twelve inbred lines, 66 diallel crosses among them and eight hybrid checks were evaluated at silage and grain harvests for 2 years at two agro-climatically diverse locations in the Federal Republic of Germany. Standard methods were used for Lab analysis of in vitro digestible organic matter (IVDOM), neutral detergent fibre (NDF), acid detergent fibre (ADF) and acid detergent lignin (ADL). In NIRS evaluation, calibration equations were developed by modified stepwise regression. The standard error of calibration was 2.5, 1.7, 1.4 and 0.4 for IVDOM, NDF, ADF and ADL, respectively. The coefficient of multiple determination was high (0.9) except for ADL. The validation statistics (standard error and correlation coefficient) were similar. In the diallel crosses, the estimates of variation (heritabilities in broad and narrow sense, genotypic and error coefficients of variation), generally, did not vary appreciably and consistently in the comparisons between Lab and NIRS methods particulary at silage harvest. Simple and rank correlations between Lab and NIRS analyses were positive and significant. These correlation coefficients based on the mean performance of the diallel crosses at silage harvest were >0.9 and at least 16 hybrids were common between the two analyses, among the upper one-third or lower one-third (22) hybrids. The study showed that NIRS analysis should be useful in maize breeding programmes wherein a large number of genotypes need to be evaluated.Abbreviations NIRS near infrared reflectance spectroscopy - Lab laboratory - IVDOM in vitro digestible organic matter - NDF neutral detergent fibre - ADF acid detergent fibre - ADL acid detergent lignin - SEC standard error of calibration - SEV standard error of validation - SD standard deviation - r simple correlation coefficient - r_s rank correlation coefficient - R² coefficient of multiple determination - h_b ² heritability in broad sense - h_n ² heritability in narrow sense - CV_g genotypic coefficient of variation - CV_e error coefficient of variation - GCA general combining ability effect     

Investigating the effects of simulated transport vibration on tomato tissue damage based on vis/NIR spectroscopy

Visible and near infrared (vis/NIR) spectroscopy combined with chemometrics were investigated to evaluate the effects of simulated transport vibration levels on damage of tomato fruit. A total of 280 tomato samples were randomly divided into 5 groups; each group was subjected to vibration at different acceleration levels. A total of 230 samples (46 from each group) were selected as a calibration set; whereas 50 samples (10 from each group) were selected as a prediction set. Raw spectra, differentiation (the first derivative) spectra, extended multiplicative scatter correction (EMSC) processed spectra and standard normal variant combined with detrending (SNV–DT) processed spectra were used for calibration models. SNV–DT processed spectra had the best performance using for partial least squares (PLS) analysis. The PLS analysis was implemented to calibrate models with different wavelength bands including visible, short-wave near infrared (SWNIR) and long-wave near infrared (LWNIR) regions. The best PLS model was obtained in the vis/NIR (600–1600 nm) region. Using a grid search technique and radial basis function (RBF) kernel, four least squares support vector machine (LS–SVM) models with different latent variables (7, 8, 9, and 10 LVs) were compared. The optimal model was obtained with 9 LVs and the correlation coefficient (r), root mean square error of prediction (RMSEP) and bias for the best prediction by LS–SVM were 0.984, 0.137 and 0.003, respectively. The results showed that vis/NIR spectroscopy could be applied as a reliable and rapid method for predicting the effect of vibration levels on tissue damage of tomato fruit.