基于Android平台微型近红外光谱仪的开发与应用

在Android平台上对C11708MA微型近红外光谱仪进行系统开发,实现光谱仪控制、样品指标测量、调用模型文件并显示样品可溶性固形物的预测结果等功能。利用近红外漫反射无损检测技术对镇江句容果园水蜜桃样品的可溶性固形物含量进行相关研究,运用化学计量学方法建立了水蜜桃可溶性固形物含量的近红外模型,并对模型的性能进行了评价。结果表明,采用偏最小二乘法(PLS)建立模型,光谱预处理的最佳条件为:移动窗口平滑(MAF)和Savitzky-Golay一阶导数。所建模型的校正相关系数(R_c)和预测相关系数(R_p)分别为0.931 1和0.880 2,校正标准偏差(RMSEC)和预测标准偏差(RMSEP)分别为0.441 0和0.531 0。开发的App程序运行稳定,预测结果准确,可应用于水蜜桃内部品质可溶性固形物含量的快速、无损、活体检测。     

蓝莓可溶性固形物、总酸近红外无损检测模型的建立

建立了近红外漫反射光谱技术检测蓝莓可溶性固形物、总酸的数学模型,并对其进行评价。实验比较了在近红外全波长范围400~2 500 nm内,不同的光谱预处理方法对模型的影响。结果表明利用偏最小二乘法(PLS)、一阶导数(D1Log(1/R))和加权多元离散校正处理(WMSC)建立的可溶性固形物含量(SSC)定标模型预测结果相对较好。其预测相关系数Rp2为0.8518,预测标准误差(SEP)为0.351,相对分析误差(RPD)为2.05。总酸的最佳模型处理条件为改进偏最小二乘法(MPLS)、二阶导数(D2Log(1/R))和WMSC,其Rp2为0.8776,SEP为0.042,RPD为2.10。由此确定近红外漫反射技术可用于蓝莓可溶性固形物、总酸含量的快速无损检测。     

䥺Symbol`@@基于近红外光谱漫透射技术的苹果可溶性固形物含量在线检测

为研究苹果的内部品质,提高检测的速度和稳定性,将近红外光谱漫透射技术应用于在线检测研究,并采取偏最小二乘回归(PLSR)算法结合不同光谱预处理方法建立苹果内部的可溶性固形物含量(SSC)的定量模型。结果表明:采用一阶微分结合多元散射校正(MSC)预处理后的模型最稳定,校正集和预测集的标准差分别为0.17和0.39,校正集的相关系数也达到0.988 3。试验结果说明近红外光谱漫透射技术能够快速、无损地检测出苹果的可溶性固形物含量。     

基于近红外光谱漫透射技术的苹果可溶性固形物含量在线检测

为研究苹果的内部品质,提高检测的速度和稳定性,将近红外光谱漫透射技术应用于在线检测研究,并采取偏最小二乘回归(PLSR)算法结合不同光谱预处理方法建立苹果内部的可溶性固形物含量(SSC)的定量模型。结果表明:采用一阶微分结合多元散射校正(MSC)预处理后的模型最稳定,校正集和预测集的标准差分别为0.17和0.39,校正集的相关系数也达到0.988 3。试验结果说明近红外光谱漫透射技术能够快速、无损地检测出苹果的可溶性固形物含量。     

基于近红外光谱技术联合极限学习机的蓝莓贮藏品质定量模型建立

采用近红外光谱技术联合极限学习机(extreme learning machine, ELM)方法建立蓝莓贮藏品质的定量检测模型,实现对蓝莓果实的可溶性固形物、维生素C和花青素含量的快速无损检测,以期为鲜食蓝莓低温贮藏期间的在线品质检测提供技术参考。利用LabSpec 5000光谱仪采集5个不同贮藏时间共150组蓝莓样本的近红外光谱,通过基于联合X/Y的异常样本识别和剔除方法筛选异常样本,使用联合X-Y距离样本集划分方法对样本集进行划分。通过对比分析标准正态变换、多元散射校正、一阶导数等预处理方法对模型性能的影响,确定蓝莓3个成分各自最优预处理方法,采用联合区间偏最小二乘算法(synergy interval partial least squares, SiPLS)选择出特征波段,将其作为输入建立ELM定量分析模型,并将模型结果与偏最小二乘回归进行对比分析。结果表明,蓝莓果实的可溶性固形物、维生素C和花青素含量最优ELM模型的校正集相关系数分别为0.920 5、0.908 7、0.942 1;验证集相关系数为0.882 6、0.897 2、0.869 3;校正集均方根误差为0.766...     

基于LVF-NIR的樱桃可溶性固形物含量无损速测模型的建立与优化

目的建立基于便携式近红外光谱仪的樱桃可溶性固形物含量无损快速定量检测模型,从而实现樱桃品质的无损快速检测。方法以北京通州产红灯樱桃、黄玉樱桃为研究对象,采用便携式线性渐变分光近红外光谱仪采集光谱数据,并采用折光仪测定其可溶性固形物含量;采用偏最小二乘回归结合全交互验证算法将光谱数据与可溶性固形物含量测定值建立定量校正模型,采用外部验证集对模型的预测性能做进一步测试。结果红灯樱桃可溶性固形物含量模型的R_C~2、RMSEC、R_(CV)~2、RMSECV、RPD分别为0.9194、0.79、0.8920、0.92、3.54,黄玉樱桃可溶性固形物含量模型的R_C~2、RMSEC、R_(CV)~2、RMSECV、RPD分别为0.8618、0.76、0.8246、0.86、2.70;两种樱桃可溶性固形物含量合并模型的R_C~2、RMSEC、R_(CV)~2、RMSECV、RPD分别为0.9125、0.81、0.8946、0.89、3.38。结论基于便携式线性渐变分光近红外光谱仪数据所建校正模型具有较好的准确度,可满足樱桃可溶性固形物含量的无损快速检测需求。     

基于Stacking框架的蓝莓硬度预测可溶性固形物含量及维生素C模型构建

目的 探索了一种基于蓝莓的质量、硬度快速、无损预测它的可溶性固形物与维生素C含量的方法，为蓝莓化学成分的预测提供一种新思路。方法 通过对蓝莓质量、硬度与可溶性固形物（soluble solid content, SSC）含量、维生素C（vitamin c, VC）相关性分析后，建立了基于一维特征的质量和硬度预测SSC与VC模型。其次，对硬度添加多项式特征做升维处理，同一维进行相同研究。最后，对比Stacking框架与单一模型，及添加多项式特征的预测效果。结果 一维特征条件下，基于Stacking框架的硬度预测SSC与VC的R2分别为0.873、0.875，预测效果优于质量与单模型预测；多维特征条件下，硬度添加到3次方时预测SSC效果最佳，R2为0.889，添加到12次方时预测VC含量效果最佳，R2为0.890，预测效果均好于一维特征。结论 Stacking框架结合添加多项式特征在蓝莓硬度快速、无损预测它的SSC及VC方面具有良好的潜力，为蓝莓品质检测提供新途径。     

不同扫描方式南果梨近红外模型差异性研究

依据近红外光谱原理,分别以不同的扫描方式对南果梨样品进行光谱扫描,并对80个南果梨样品分别建立可溶性固形物(SSC)、有效酸度(pH)模型,模型的相关系数和校正集标准偏差均达到应用要求,应用所建模型对20个已知成分含量的南果梨进行可溶性固形物和有效酸度的预测,预测值方差分析结果表明模型间有显著差异,确定180°转动扫描2次(正反两面扫描两次)条件下建立的模型为最佳.     

紫外/可见光谱技术无损检测李子可溶性固形物含量

利用光谱技术结合化学计量学对李子可溶性固形物含量检测进行研究,为李子品质无损检测提供科学方法。通过反射式光谱采集系统获取了"红"李子和"青"李子的平均光谱,并对原始光谱数据进行预处理;应用连续投影算法(SPA)和竞争性自适应重加权算法(CARS)对预处理后的光谱数据提取特征波长;分别建立基于全光谱和特征波长的预测李子可溶性固形物含量的误差反向传播(BP)网络模型。结果表明:利用SPA和CARS算法分别从全光谱的1024个波长中选取出31和104个特征波长;而基于特征波长建立的CARS-BP网络模型效果最优,其相关系数r_c为0.998,r_p为0.887,均方根误差RMSEC为0.026,RMSEP为1.767。这表明光谱技术结合化学计量学进行李子可溶性固形物含量的无损检测具有可行性。     

近红外光谱技术快速无损检测蓝莓总黄酮、花青素的研究

为建立起蓝莓的快速无损检测体系,本研究应用近红外光谱技术对鲜蓝莓中总黄酮、花青素的含量进行了分析。在光谱全波长(400~2500 nm)范围内采用偏最小二乘法(PLS)建立蓝莓总黄酮、花青素含量的定标数学模型,其相关系数分别为0.836和0.750,校正标准误差(SECV)分别为1.423 mg/(100g)和4.688 mg/(100g)。然后使用最优模型对32个未知样品进行预测,其预测相关系数Rp2分别为0.7968和0.7902,预测标准误差分别为2.779 mg/(100g)和5.013 mg/(100g),残差和分别为-0.003 mg/(100g)和-9.256(mg/100g)。实验结果说明,近红外漫反射技术可用于快速无损检测蓝莓中总黄酮、花青素含量。     

Maturity sorting index of dragon fruit: Hylocereus polyrhizus

The purpose of this research was to investigate maturity prediction of red flesh dragon fruit based on non-destructive measures. Specific weight, sphericity, color value L, a, b and light reflectance spectrum were linearly combined by partial least squares regression (PLSR) analysis. The PLSR models could predict days after fruit set, weight ratio and total soluble solids relatively well with standard deviation divided by standard error of prediction (RPD) of 2.86, 2.45 and 2.38, respectively. Date after fruit set, total soluble solids, total acid, ratio of total soluble solids and total acidity and weight ratio were transformed into a principal component 1 (PC1) by the principal component analysis and used to represent a single maturity index. The PLSR model with non-destructive parameters resulted in an improved performance in the prediction of the maturity index (PC1) with a RPD increase to 3.49. The model could be further simplified but retained a comparable accuracy by the application of a log (R680/R550) in place of the light reflectance spectrum.     

Measurement of the degree of starch damage in flour by near infrared reflectance analysis

The degree of mechanical damage imparted to starch granules during flour milling is of importance in relation to the baking properties of the flour. A calibration is presented which enables the degree of starch damage to be measured by near infrared reflectance. This technique is rapid and non-destructive so facilitating quality control of a flour mill. The results obtained using 148 UK commercial flour samples were a correlation coefficient of 0.906 and a residual standard deviation of 3.5 Farrand units using 74 of the samples covering the range 6-47 Farrand units. The calibration was obtained using log (1/R) data at 1442, 1580, 2060 and 2258 nm. The residual standard deviation increased to 4.2 Farrand units when the calibration was used to predict the values of 68 further samples of commercial flour. The wavelengths at which reflectance measurements were taken corresponded to overtones and combinations of vibration frequencies due to free and hydrogen-bonded O—H bonds in starch.     

基于近红外高光谱成像快速无损检测注胶肉研究

采用近红外高光谱成像技术结合化学计量学方法建立注胶肉的快速无损检测模型。首先通过近红外高光谱成像系统获取含有不同浓度梯度卡拉胶的猪里脊肉高光谱图像,然后提取图像中的光谱数据,使用偏最小二乘法(Partial least square,PLS)探究光谱信息与不同掺假比例卡拉胶之间的定量关系。结果表明全波段光谱(900~1700 nm)所构建的PLS校正集模型均方根误差(Root mean square error,RMSE)为1.74%,预测模型RMSE为3.16%。表明基于全波段所建立的PLS模型具有较优的预测性能。利用连续投影算法(Successive projection algorithm,SPA)筛选获得11个特征波长,并优化全波长PLS模型,将预测集样品带入,以验证模型的预测效果,结果表明SPA算法结合PLS建模方法所建立的模型预测效果更优,预测集相关系数(R_P)为0.93,均方根误差(Root mean square error of prediction,RMSEP)为3.51%,预测偏差(Residual predictive deviation,RPD)为2.66。试验表明利用高光谱成像技术可实现对注胶猪肉的快速无损检测。     

京郊鲜食杏白利糖度的便携式光谱快速无损检测方法研究

目的 建立京郊鲜食杏白利糖度的定量分析预测模型, 实现对京郊鲜食杏品质的快速无损检测。方法 使用便携式近红外光谱仪采集900~1700 nm下鲜食杏的漫反射光谱信息, 使用多元散射校正(multiplicative scatter correction, MSC)、标准正态变量变换(standard normal variable transformation, SNV)和Savitzky-Golay卷积平滑(Ssavitzky – Ggolay smooth, S-G)对原始光谱数据进行预处理, 使用Kennard-Stone (K-S)算法以3:1比例将样本集划分成校正集和预测集, 利用竞争自适应重加权采样(competitive adaptive reweighted sampling, CARS)算法和连续投影算法(successive projections algorithm, SPA)对光谱进行特征波长筛选, 使用偏最小二乘回归(partial least squares regression, PLSR)算法建立京郊鲜食杏白利糖度的预测模型。结果 以MSC+S-G+ CARS+PLSR算法建立的北京鲜食杏的白利糖度预测模型取得较好的预测精度, 模型的校正集均方根误差、校正集相关系数、预测集均方根误差、预测集相关系数分别为0.3502、0.9747、0.4698、0.9616。结论 基于便携式近红外光谱技术所采集数据构建的京郊鲜食杏白利糖度预测模型准确性较高, 可以快速准确检测鲜食杏白利糖度, 从而实现对鲜食杏品质的快速无损检测, 为鲜食杏的品质检测提供了理论依据和方法指导。     

樱桃多品质数据分析与无损快速检测模型的建立

目的 研究樱桃多品质数据分布情况, 建立樱桃多品质无损快速检测方法。方法 对樱桃样品分别测试可溶性固形物含量、可滴定酸含量、果实硬度。采用统计分析方法对数据进行统计学描述, 分别绘制含量分布直方图并计算直方图分布频次百分比。以樱桃样品近红外光谱数据为自变量、品质数据参考值为因变量建立樱桃品质无损快速定量检测模型。结果 统计分析结果表明, 可溶性固形物含量11~17 Brix区间范围内的样品数占样品总数的约86.0%, 可滴定酸含量0.1%~0.8%区间范围内的样品数占样品总数的约90.4%, 果实硬度1.60~3.00 kg/cm2区间范围内的样品数占样品总数的约86.0%。多元回归建模结果表明, 剔除异常值有助于提高模型预测性能, 剔除异常值后可溶性固形物含量、可滴定酸含量、果实硬度模型的相对预测性能值分别提高了15.3%、32.9%、12.3%。结论 采用统计分析结合直方图分析可较直观地描述樱桃品质分布情况; 剔除异常值对提高樱桃可滴定酸含量近红外无损检测模型预测能力的作用最大。     

基于近红外光谱建立纸张表面施胶量测定模型

采集不同施胶量纸张的近红外光谱,利用偏最小二乘法建立测定纸张表面施胶量基于近红外光谱的校正模型。得到校正模型的交叉验证均方差（RMSECV）和外部验证均方差（RMSEP）分别为0.0928和0.1460,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为0.9609和0.9294,表明所建立的校正模型具有较高的预测精度和较好的推广性,为纸张无损伤、无预处理的快速、简便、准确的检测提供了新的途径,并且可望实现纸机上的在线检测。     

Non-destructive evaluation of maturity and quality parameters of pomegranate fruit by visible/near infrared spectroscopy

In this study, the potential of visible and near infrared spectroscopy was investigated to classify the maturity stage and to predict the quality attributes of pomegranate variety “Ashraf” such as total soluble solids content, pH, and titratable acidity during four distinct maturity stages between 88 and 143 days after full bloom. Principal component analysis was used to distinguish among different maturities. The prediction models of internal quality attributes of the pomegranate were developed by partial least squares regression. The transmission spectra of pomegranate were obtained in the wavelength range from 400 to 1100 nm. In this research several preprocessing methods were utilized including centering, smoothing (Savitzky–Golay algorithm, median filter), normalization (multiplicative scatter correction and standard normal variate) and differentiation (first derivative and second derivative). It concluded that different preprocessing techniques had effects on the classification performance of the model using the principal component analysis method. In general, standard normal variate and multiplicative scatter correction gave better results than the other pretreatments. The correlation coefficients (r), root mean square error of calibration and ratio performance deviation for the calibration models were calculated: r = 0.93, root mean square error of calibration = 0.22 °Brix and ratio performance deviation = 6.4 °Brix for total soluble solids; r = 0.84, root mean square error of calibration = 0.064 and ratio performance deviation = 4.95 for pH; r = 0.94, root mean square error of calibration = 0.25 and ratio performance deviation = 5.35 for titratable acidity.     

Development of a multispectral imaging system for online quality assessment of pomegranate fruit

The objective of this study was to develop a prototype multispectral imaging system for online quality assessment on pomegranate fruit. At first, a visible/near infrared spectroscopy (400–1100 nm) was tested for non-destructive determination of total soluble solids, titratable acidity, and pH. The spectral data were analyzed using the partial least square analysis. Then to establish consistent multispectral imaging system, the highest absolute values of β-coeffcients correspond to wavelengths from the best partial least square calibration model were selected and used for identifying the optimal wavelengths. Consequently, a multispectral imaging system was developed based on the effective wavelengths 700, 800, 900, and 1000 nm. The performance of the developed multispectral imaging system was evaluated by multiple linear regression models. The multiple linear regression model predict total soluble solids with r = 0.97, root mean square error of calibration = 0.21°Brix, and ratio performance deviation = 6.7 °Brix. Also, the results showed that the models had good predictive ability for pH and titratable acidity. Results showed that the developed multispectral imaging system based on the optimal wavelengths could be used for online quality assessment of pomegranate fruit.     

近红外法测定豆浆蛋白质、脂肪和可溶性固形物含量

利用傅里叶变换近红外光谱仪采用积分球漫反射方式对60个豆浆样品进行光谱的采集,结合常规分析结果分别建立了3种成分的近红外校正模型。结果表明:豆浆蛋白质、脂肪及可溶性固形物光谱分别经过消除常数偏移量、一阶导数和矢量归一化(SNV)预处理后建模效果最好。蛋白质、脂肪和可溶性固形物含量的校正模型决定系数(R2)分别为:0.966 4、0.950 0和0.950 7,交叉验证均方根差(RMSECV)依次为0.076 9、0.087 4和0.316;对模型进行外部验证,验证集化学值和模型预测值之间差异不显著,说明模型可以用于豆浆中蛋白质、脂肪和可溶性固形物含量的检测。     

Mapping the Pungency of Green Pepper Using Hyperspectral Imaging

The pungency level of green peppers is dependent on the amounts of capsaicin and dihydrocapsaicin they contain. This study was conducted to develop a non-destructive method for the prediction and mapping of the capsaicin and dihydrocapsaicin contents in green pepper. Hyperspectral images of 200 total green peppers of three varieties were acquired in the wavelength range of 1000–1600 nm, from which the mean spectra of each pepper variety were extracted. The reference capsaicin and dihydrocapsaicin contents of the samples were measured by high-performance liquid chromatography. Quantitative calibration models were built using partial least squares (PLS) regression with different spectral preprocessing techniques; the best performance was found by normalizing the preprocessed spectra with correlation coefficients (r_pred) of 0.86 and 0.59, which showed the standard errors of prediction (SEPs) of 0.09 and 0.03 mg/g for capsaicin and dihydrocapsaicin, respectively. Seventeen and 16 optimal wavebands were selected using the successive projections algorithm; r_pred of 0.88 and 0.68 and SEPs of 0.084 and 0.027 mg/g were obtained for capsaicin and dihydrocapsaicin, respectively, from the newly developed PLS calibration models using these optimal wavebands. The successive projections algorithm (SPA)-PLS model was used to map the capsaicin and dihydrocapsaicin contents of the green peppers. These maps provided detailed information on the pungency levels of the tested green peppers. The results of this study indicated that hyperspectral imaging is useful for the rapid and non-destructive evaluation of the pungency of green peppers.