基于BP神经网络检测面粉中滑石粉含量的研究

利用近红外光谱技术对掺杂滑石粉的小麦面粉进行了检测,采用多元散射校正对谱图进行预处理,利用BP神经网络中的SCG反向传播算法训练函数建立了面粉中滑石粉的定量分析模型,并对校正集和预测集进行了定量分析,分析结果为R2=0.997 3,RMSEC=0.436 7,RMSEP=1.708 8.结果表明,BP神经网络结合近红外光谱技术检测面粉中滑石粉含量具有快速、精度高、泛华能力强的优点,可用于面粉中滑石粉含量的快速准确检测.     

榆树木材基本密度近红外模型优化的研究

 为探究近红外光谱技术野外测量木材基本密度的可行性,用圆盘模拟伐倒木锯面,采集光谱信号,结合偏最小二乘法（PLS）建立榆树木材基本密度预测模型.其校正模型和验证模型决定系数R²分别为0.8456和0.8011,均方根误差RMSE分别为0.0231和0.0266,标准误差SE分别为0.0232和0.0268.为进一步提高模型预测精度,利用卷积平滑、小波变换等6种方法对光谱信号进行预处理.结果表明,基于小波变换去噪的模型精度最好,校正模型和验证模型决定系数分别为0.8996和0.8662,RMSE和SE的值均达到最小.研究表明,近红外光谱技术可用于木材基本密度的野外测量.     

近红外光谱-偏最小二乘法测定三元混合物中甲醇和乙醇的含量

应用偏最小二乘法建立了甲醇、 乙醇和水三元混合体系近红外光谱的校正模 型, 采用交互验证方式对模型精度进行检验. 通过选取波长, 使所建模型中甲醇测定相关 系数r达到0.999　91, 交互验证均方根误差(RMSECV)为0.431, 乙醇的r 达到0.999 98, RMSECV为0.193.　用所建模型测定样品, 与气相色谱法 分析结果相近, 相对误差小于3.505％.     

用近红外光谱法实时监测蛹虫草发酵中胞内多糖的质量浓度

应用近红外光谱(NIR)结合偏最小二乘法(PLS)建立一种实时监测蛹虫草发酵中胞内多糖质量浓度的新方法.对39个批次的蛹虫草在3个不同条件的5L发酵罐中进行蛹虫草深层发酵,发酵过程中间隔一定时间取样,采集样品的近红外光谱,并按常规方法测定样品中胞内多糖质量浓度,再采用PLS法建立样品的近红外光谱与胞内多糖质量浓度间的模型,所建模型经过选择最适光谱预处理方法和最适隐变量数进行优化,其留一交互验证预测值与化学测定参考值间的相关系数R=0.8750,交互验证均方根误差RMSECV=0.3052.采用最优PLS模型对样品中胞内多糖质量浓度进行预测,校正集预测均方根误差RMSEC=0.1670,预测集预测均方根误差RMSEP=0.3650,表明模型的稳健性和预测性能较好。     

近红外光谱在印刷品颜色检测中的应用

为了对印刷品颜色进行快速、准确检测,应用近红外光谱技术(NIR)并结合偏最小二乘法(PLS)建立印刷品颜色检测模型.对近红外光谱获取的144个样本光谱曲线,应用主成分分析方法进行降维,维数为5.选取的主成分作为光谱优化特征子集以替代原来复杂的光谱数据.随后,将144个样本数据随机分为定标集和预测集,利用偏最小二乘法在103个定标集样本数据基础上建立印刷品颜色预测模型,应用此模型对41个预测集样本颜色进行预测.研究结果表明:利用PLS模型得到样本的实测值和预测值之间的预测决定系数(R~2)为99.74%,预测平均相对误差为0.636%,表明利用近红外光谱技术检测印刷品颜色是可行的.     

基于降噪处理的蒙古栎木材气干密度NIRS定标模型

分别采用卷积平滑法、小波变换法对蒙古栎木材近红外光谱(NIRS)做去噪处理,并讨论两者混合去噪时,处理顺序变化对光谱去噪效果的影响,最后应用偏最小二乘法(partial least squares regression,PLS)和主成分回归法建立蒙古栎木材气干密度近红外定标模型。结果表明,当平滑点数为3,db5小波分解层数为2时,以平滑+小波方式去噪效果最好,其信噪比(SNR)为18.546,均方根误差为0.04。平滑+小波去噪后,基于PLS的蒙古栎木材密度近红外校正模型决定系数由0.767提高到0.902,校正均方根误差降低了35.32%,预测集决定系数为0.860,内部交叉验证和预测均方根误差分别达到最低,剩余预测偏差为2.67。因此,近红外光谱技术可实现蒙古栎木材气干密度快速预测,合理选择处理参数和建模方法可以有效提高模型精度。     

可见-短波近红外光谱用于血红蛋白分析及其稳定性

基于随机性、相似性和稳定性,通过定标集、预测集、检验集的建模过程,采用可见-近红外(NIR)光谱结合偏最小二乘(PLS)方法建立人类溶血液样品的血红蛋白(Hb)的分析模型。将全谱扫描区(400—249 8 nm)分成可见区(400—780nm)、短波近红外区(780—110 0 nm)、长波近红外区(1100—249 8 nm)、可见-短波近红外区(400-1100 nm)、全近红外区(780—249 8 nm)。经过比较、检验,结果表明,可见-短波近红外区达到了最好的模型效果和稳定性,最优PLS因子数为7,检验的预测均方根误差(V-SEP)和预测相关系数(V-RP)分别为4.42 g.L-1、0.967,达到了高的预测精度和稳定性。     

多维标度和随机森林应用于汽油近红外定量分析

随机森林(RF)回归应用于汽油辛烷值的近红外定量模型的波长优选。提出的双中心指标多维标度(DC-MDS)方法能够有效地找到定标和预测样品集的合理划分。RF回归建模的过程中选择采用较大的决策树数量(nTree=500),避免建模过程发生拟合,进一步调试并选择最优的分裂变量数(mtry=130);最后在最优参数的RF建模过程中提取具有最大贡献的30个信息波长,为汽油辛烷值的测定建立离散波长的近红外定量分析模型;其预测决定系数为0.971,预测均方根偏差为0.219%。结果表明,RF回归具有应用于汽油辛烷值近红外定量测定的潜力。     

铜掺杂二氧化硅用于雪菊中木犀草苷的富集及近红外光谱分析

用制备的铜掺杂二氧化硅作为吸附剂, 富集雪菊提取液中的木犀草苷, 测量富集有目标组分吸附剂的近红外光谱, 所得光谱经预 处理后, 用偏最小二乘法建立木犀草苷的定量校正模型并进行验证. 结果表明： 当铜掺杂二氧化硅的用量为0.25 g、 常温振荡20 min时, 对木犀草苷吸附率达89.7%； 雪菊提取液中的木犀草苷经吸附剂富集后, 无需脱附可直接检测； 所得近红外光谱经多元散射校正结合一阶导数预处理后, 木犀草苷校正模型的预测质量浓度和参考质量浓度间的相关系数为0.975 0, 交叉验证均方根误差为0.959 8 mg/L, 木犀草苷在1.5~19.5 mg/L的较低质量浓度范围内, 预测集的回收率可达82.6%~111.4%.     

Analysis of the global T_m–T_s correlation and establishment of the latitude-related linear model

In this study,the correlation between Tm,a key variable in GNSS water vapor inversion,and surface temperature(Ts)was calculated on a global scale based on the global geodetic observing system(GGOS)atmosphere Tmdata and European centre for medium-range weather forecasts(ECMWF)surface temperature data.The results show that their correlation is mainly affected by latitudes,and the correlation is stronger at high latitudes and weaker at low latitudes.Although the correlation is relatively weak in the tropic areas,the temperature changes so little in a year in these areas that we can still achieve good Tmresults by linear regression model.Based on these facts,‘‘GGOS atmosphere’’Tmdata and ECMWF Tsdata from 2005 to2011 were used to establish the global latitude-related linear regression model.The new model has root mean square error(RMSE)of 3.2,3.3,and 4.4 K,respectively,compared with respect to the‘‘GGOS atmosphere’’data,COSMIC data,and radiosonde data and is more accurate than the Bevis Tm–Tsrelationship.     

铜掺杂二氧化硅用于雪菊中木犀草苷的富集及近红外光谱分析

用制备的铜掺杂二氧化硅作为吸附剂, 富集雪菊提取液中的木犀草苷, 测量富集有目标组分吸附剂的近红外光谱, 所得光谱经预 处理后, 用偏最小二乘法建立木犀草苷的定量校正模型并进行验证. 结果表明： 当铜掺杂二氧化硅的用量为0.25 g、 常温振荡20 min时, 对木犀草苷吸附率达89.7%; 雪菊提取液中的木犀草苷经吸附剂富集后, 无需脱附可直接检测; 所得近红外光谱经多元散射校正结合一阶导数预处理后, 木犀草苷校正模型的预测质量浓度和参考质量浓度间的相关系数为0.975 0, 交叉验证均方根误差为0.959 8 mg/L, 木犀草苷在1.5~19.5 mg/L的较低质量浓度范围内, 预测集的回收率可达82.6%~111.4%.     

巴戟天药材中耐斯糖含量近红外光谱测定方法的建立

目的建立巴戟天药材中耐斯糖含量的近红外光谱测定方法。方法用高效液相色谱法测定114批巴戟天药材中耐斯糖含量,采集近红外光谱后,运用多元散射校正法,结合最小偏二乘法建立巴戟天耐斯糖含量的定量模型。结果建立的耐斯糖近红外光谱定量模型,内部交叉验证决定系数为0.979 1,校正标准偏差为0.909 0,预测标准差为0.909 0,交叉验证的标准偏差为1.093 0。结论该含量测定近红外光谱模型稳定、准确适用于巴戟天药材中耐斯糖的含量测定。     

Determination of soluble solid content and acidity of loquats based on FT-NIR spectroscopy

The near infrared (NIR) spectroscopy technique has been applied in many fields because of its advantages of simple preparation, fast response, and non-destructiveness. We investigated the potential of NIR spectroscopy in diffuse reflectance mode for determining the soluble solid content (SSC) and acidity (pH) of intact loquats. Two cultivars of loquats (Dahongpao and Jiajiaozhong) harvested from two orchards (Tangxi and Chun’an, Zhejiang, China) were used for the measurement of NIR spectra between 800 and 2500 nm. A total of 400 loquats (100 samples of each cultivar from each orchard) were used in this study. Relationships between NIR spectra and SSC and acidity of loquats were evaluated using partial least square (PLS) method. Spectra preprocessing options included the first and second derivatives, multiple scatter correction (MSC), and the standard normal variate (SNV). Three separate spectral windows identified as full NIR (800~2500 nm), short NIR (800~1100 nm), and long NIR (1100~2500 nm) were studied in factorial combination with the preprocessing options. The models gave relatively good predictions of the SSC of loquats, with root mean square error of prediction (RMSEP) values of 1.21, 1.00, 0.965, and 1.16 °Brix for Tangxi-Dahongpao, Tangxi-Jiajiaozhong, Chun’an-Dahongpao, and Chun’an-Jiajiaozhong, respectively. The acidity prediction was not satisfactory, with the RMSEP of 0.382, 0.194, 0.388, and 0.361 for the above four loquats, respectively. The results indicate that NIR diffuse reflectance spectroscopy can be used to predict the SSC and acidity of loquat fruit.     

近红外光谱结合偏最小二乘法测定福多斯坦成分含量

采用近红外漫反射光谱分析技术和偏最小二乘法对福多斯坦药物的有效成分进行定量分析测定,采集48个不同比例的福多斯坦样品近红外漫反射光谱,用一阶导数的光谱预处理方法,结合偏最小二乘法建立福多斯坦的定量分析模型.结果显示:交互验证均方根误差为0.003 57,相关系数R为0.994 77,预测均方根误差为0.003 89,平均回收率为99.63%(n=8),结果表明,用近红外光谱分析技术联合偏最小二乘法对福多斯坦进行定量分析结果准确可靠,方法简便快速.     

Incorporating temporal variability to improve geostatistical analysis of satellite-observed CO2 in China

Observations of atmospheric carbon dioxide (CO2 ) from satellites offer new data sources to understand global carbon cycling. The correlation structure of satellite-observed CO2 can be analyzed and modeled by geostatistical methods, and CO2 values at unsampled locations can be predicted with a correlation model. Conventional geostatistical analysis only investigates the spatial correlation of CO2 , and does not consider temporal variation in the satellite-observed CO2 data. In this paper, a spatiotemporal geostatistical method that incorporates temporal variability is implemented and assessed for analyzing the spatiotemporal correlation structure and prediction of monthly CO2 in China. The spatiotemporal correlation is estimated and modeled by a product-sum variogram model with a global nugget component. The variogram result indicates a significant degree of temporal correlation within satellite-observed CO2 data sets in China. Prediction of monthly CO2 using the spatiotemporal variogram model and spacetime kriging procedure is implemented. The prediction is compared with a spatial-only geostatistical prediction approach using a cross-validation technique. The spatiotemporal approach gives better results, with higher correlation coefficient (r2 ), and less mean absolute prediction error and root mean square error. Moreover, the monthly mapping result generated from the spatiotemporal approach has less prediction uncertainty and more detailed spatial variation of CO2 than those from the spatial-only approach.     

PCA结合马氏距离法剔除近红外异常样品

采用PCA(principal component analysis)结合马氏距离法对近红外校正样品集中的异常样品进行剔除,从校正集的60个食醋样品中剔除了12个异常样,用剩下的48个样品建立了总酸、挥发酸的校正模型,并对预测集的15个食醋样品进行预测分析,以相关系数(R)、预测均方差(RMSEP)、平均相对误差(Er)为预测模型的评价指标.预测集R分别为0.9759,0.9383;RMSEP分别为0.0938,0.1635;Er分别为1.34%,2.80%.与原始校正集所建模型相比,校正模型的预测精度和稳定性得到显著提高.     

Quantifying biochemical variables of corn by hyperspectral reflectance at leaf scale

To further develop the methods to remotely sense the biochemical content of plant canopies, we report the results of an experiment to estimate the concentrations of three biochemical variables of corn, i.e., nitrogen （N）, crude fat （EE） and crude fiber （CF） concentrations, by spectral reflectance and the first derivative reflectance at fresh leaf scale. The correlations between spectral reflectance and the first derivative transformation and three biochemical variables were analyzed, and a set of estimation models were established using curve-fitting analyses. Coefficient of determination （R2）, root mean square error （RMSE） and relative error of prediction （REP） of estimation models were calculated for the model quality evaluations, and the possible opti- mum estimation models of three biochemical variables were proposed, with R2 being 0.891, 0.698 and 0.480 for the estimation models of N, EE and CF concentrations, respectively. The results also indicate that using the first derivative reflectance was better than using raw spectral reflectance for all three biochemical variables estimation, and that the first derivative reflectances at 759 nm, 1954 nm and 2370 nm were most suitable to develop the estimation models of N, EE and CF concentrations, respectively. In addition, the high correlation coefficients of the theoretical and the measured biochemical parameters were obtained, especially for nitrogen （r=0.948）.     

近红外光谱定量检测丙烷和异丁烷

近红外光谱分析技术结合偏最小二乘法（PLS）对多成分挥发性有机物（VOCs）进行连续的在线监测具有重要意义．用傅里叶变换红外光谱仪（FTIR）分析了丙烷和异丁烷2种挥发性有机物的近红外光谱特征．采用线性回归建模方法——偏最小二乘法在丙烷和异丁烷混合气体的近红外光谱范围（5600～6200cm^-1）内建立了预测模型．基于该模型预测了验证集样品中2种气体的含量,并对模型进行了评价．结果表明,对2种气体浓度的预测比较准确,相对误差基本在5％以内．     

近红外光谱-偏最小二乘法无损定量分析异烟肼片

应用近红外漫反射光谱结合偏最小二乘法， 对异烟肼片中异烟肼的含量进行分析， 建立了近红外光谱数学校正定量分析模型， 其对校正集样品的交互验证均方根误差(RMSECV)为0.00632. 对预测集样品的预测均方根误差(RMSEP)为0.00603； 回归系数为0.99456;加样平均回收率为99.772%. 重现性实验相对标准偏差(RSD)为0.526%. 结果表明， 该方法预测精度高， 且具有方便快捷、 非破坏、 无污染、 可在线检测和重现性好等优点.     

泛化误差的三种交叉验证估计方法的比较

在泛化误差（GeneralizationError）的估计中,交叉验证（Cross—validation）是最常用的方法．基于均方误差准则下,采用生物信息数据比较了泛化误差的5折、10折交叉验证和组块3×2交叉验证估计,实验证明组块3×2交叉验证比5折、10折交叉验证方法更好．