基于改进离散粒子群算法的青贮玉米原料含水率高光谱检测

快速、无损和准确检测青贮玉米原料含水率,对确保青贮玉米发酵品质、推动青贮产业健康快速发展有重要现实意义。为探究高光谱技术在青贮玉米原料含水率检测方面的可行性,研究通过高光谱成像系统获取青贮玉米原料高光谱图像并利用烘箱加热法测定实际含水率。在粒子更新方式和惯性权重2个方面对传统离散粒子群算法(discretebinary particle swarm optimization,DBPSO)进行优化,提出基于改进型离散粒子群算法(modified discrete binary particle swarm optimization,MDBPSO)的特征波段优选方法,并利用相关系数分析法(correlation coefficient,CC)、DBPSO和MDBPSO法提取原料含水率高光谱特征变量,基于全波段反射光谱(total spectral reflectance,TSR)和特征波段反射光谱建立青贮玉米原料含水率预测模型。结果表明,MDBPSO优选特征波段适应度函数的收敛精度和收敛效率较DBPSO法均有明显改善,最优适应度值由0.761 6提高至0.812 3,函数收敛迭代次数由280次降低至79次。MDBPSO-PLSR预测模型的建模精度和预测精度均高于CC-PLSR、DBPSO-PLSR和TSR-PLSR预测模型,其校正集决定系数Rc2和均方根误差RMSEC(root mean square error of calibration)分别为0.81和0.032,预测集决定系数Rp2和均方根误差RMSEP(root mean square error of prediction)分别为0.80和0.045。该研究表明,利用高光谱图像技术检测青贮玉米原料含水率具有较高的精度,研究可为后续开发青贮玉米原料水分快速检测仪器提供借鉴方法。     

拉曼光谱法无损检测蜂蜜中的果糖和葡萄糖含量

应用拉曼光谱结合化学计量学方法对蜂蜜果糖和葡萄糖含量进行了定量分析。用自适应迭代重加权惩罚最小二乘(adaptive iteratively reweighted penalized least squares,airPLS)算法进行基线校正,用竞争性自适应重加权采样(competitive adaptive reweighted sampling,CARS)算法筛选变量,分别用线性的偏最小二乘(partial least squares,PLS)回归算法和非线性的支持向量机(support vector machines,SVM)回归算法建立定量校正模型,并进行预测。2种模型都有较好的预测结果。对果糖,SVM模型预测值与高效液相色谱法(high performance liquid chromatography,HPLC)测定值的相关系数(R)和预测均方根误差(root mean square error of prediction,RMSEP)分别为0.902和1.401,略优于PLS模型(R为0.892,RMSEP为1.604);对葡萄糖,PLS模型的R和RMSEP分别为0.968和0.669,优于SVM模型(R为0.933,RMSEP为1.410)。结果表明拉曼光谱结合化学计量学方法可快速无损测定蜂蜜果糖和葡萄糖含量。     

基于可见/短波近红外光谱检测结球甘蓝维生素C含量

维生素C是人类必需的营养素,结球甘蓝作为主要蔬菜品种之一富含维生素C。该试验利用可见/短波近红外光谱分析技术,开展结球甘蓝维生素C含量的快速检测方法研究。首先通过Kennard-Stone(K-S)法将样本按照6:1划分为校正集(60个样本)和验证集(11个样本),分别利用反射率和吸光度的原始光谱、一阶导数(first derivative,FD)和二阶导数(second derivative,SD)光谱预处理后对应的6个数据集,建立偏最小二乘(partial least squares,PLS)回归模型。针对最优光谱预处理方法处理后的光谱,设置5个置信水平(0.95,0.975,0.99,0.995,0.999 5),利用逐步回归(stepwise regression,SR)进行建模波长选择,以各置信水平对应的各组优选波长进行多元线性回归建模。结果表明:利用FD光谱预处理方法可以提高PLS回归模型精度,验正集R~2从处理前的0.85提高到0.96,是该研究的最佳光谱数据预处理方法。利用降维后的7个主成分继续建立PLS回归模型,校正集R~2为0.92,交互验证均方根误差(root mean squared error of cross validation,RMSECV)为0.658 0 mg/100 g,验证集R~2为0.96,预测均方根误差(root mean squared error of prediction,RMSEP)为1.620 4 mg/100 g。PLS回归模型预测维生素C含量,检测精度高,可以代替传统检测方法,为结球甘蓝的品质评定提供一种新的途径。进一步分别应用8,6,5个优选波长进行多元线性回归建模,校正集R~2平均为0.78,RMSECV平均为3.760 9 mg/100 g,验证集R~2平均为0.73,RMSEP平均为2.879 2 mg/100 g,虽然R~2有所降低,但波长点少,利用较少的波长变量来预测维生素C含量,降低模型复杂度,可以为便携式检测仪器开发提供技术支持,以提高结球甘蓝内部品质评定作业效率。     

结合高光谱信息的土壤有机碳密度地统计模型

传统线性回归模型在借助光谱信息进行土壤属性预测时,通常忽略了土壤自身所具有的空间异质性和依赖性,并且未考虑模型残差的空间结构。针对以上不足,该文以江汉平原232个土壤样本为研究对象,以土壤反射光谱为辅助变量,采用偏最小二乘回归、普通克里格、协同克里格以及回归克里格分别构建土壤有机碳密度预测模型,选取决定系数(R~2)、均方根误差、标准差与预测均方根误差比(ratio of performance to deviation,RPD)对模型预测精度进行对比评价。结果显示,结合高光谱信息,且同时考虑残差空间结构的回归克里格模型表现优于其他模型,预测决定系数R~2为0.617,RPD为1.614。鉴于土壤光谱信息同时还具有测定简单、省时、无损等优点,因此土壤光谱是土壤有机碳密度空间插值的理想辅助因子。     

去除水分影响提高土壤有机质含量高光谱估测精度

土壤水分的影响是当前采用光谱分析法预测土壤养分含量的关键问题,该文旨在探索去除土壤水分影响、提高有机质高光谱定量估测精度的方法。首先采用地物光谱仪进行湿土和过筛干土的高光谱测试,并进行一阶导数变换;然后,采用奇异值分解(singular value decomposition,SVD)结合相关分析筛选土壤水分特征光谱,构建去除水分因素的修正系数,形成湿土光谱的校正光谱;最后基于校正前后湿土光谱,应用偏最小二乘(partial least squares,PLS)回归构建土壤有机质含量的估测模型,并对模型进行验证和比较,分析评价校正前后光谱的预测精度。结果显示:按土壤水分含量梯度划分的2组和全部棕壤及褐土土样共4组样本校正后建模决定系数和均方根误差分别为0.85、0.82、0.74、0.76和0.19%、0.20%、0.23%、0.19%,决定系数提高了0.02~0.09,均方根误差降低了0.01~0.03百分点,验证决定系数、均方根误差和相对分析误差分别为0.78、0.77、0.72、0.76,0.21%、0.15%、0.21%、0.15%和2.03、2.02、1.86、1.98,决定系数提高了0.06~0.15,均方根误差除褐土土样提高0.02百分点外,其他样本组降低了0.01~0.08百分点,相对分析误差提高了0.17~0.43,模型决定系数和相对分析误差得到显著提升;尤其对于土壤水分含量变异系数较小的3组土样,模型从待改进级别提高到性能良好级别,对土壤有机质含量具有较好的预测准确性。说明该方法用于去除土壤水分因素影响和提高有机质含量高光谱估测精度的有效性。     

玉米作物系数无人机遥感协同地面水分监测估算方法研究

该文研究不同水分胁迫条件下无人机遥感与地面传感器协同估算玉米作物系数的可行性。利用自主研发的六旋翼无人机遥感平台搭载多光谱传感器获取内蒙古达拉特旗昭君镇试验站不同水分胁迫下大田玉米冠层光谱影像,计算植被指数,采用经气象因子和作物覆盖度校正后的FAO-56双作物系数法计算玉米的作物系数,研究作物系数与简单比值植被指数(simple ratio index,SR)、叶面积指数(leaf area index,LAI)和表层土壤含水率(surface soil moisture,SM)的相关关系,结果表明,作物系数与SR、LAI和SM的相关程度与水分胁迫程度有关,但均呈现出显著或极显著的线性关系,说明了基于这些指标建立作物系数估算模型的可能性。利用逐步回归分析方法建立了作物系数的估算模型,其估算模型,修正的决定系数、均方根误差和归一化的均方根误差分别为0.63、0.21、25.16%。经验证,模型决定系数、均方根误差和归一化的均方根误差分别为0.60、0.21、23.35%。研究结果可为利用无人机多光谱遥感平台进行作物系数估算提供技术参考。     

基于线性回归的玉米生物量预测模型及验证

玉米生物量是评估玉米长势的重要参数,为了实现玉米生物量的快速测量,该文拟以玉米株高H、茎粗长轴L、茎粗短轴S为输入,建立玉米生物量鲜质量FW和干质量DW的预测模型。采用多元回归和逐步回归方法对平展型和紧凑型玉米的小喇叭口期生物量数据进行线性回归分析。结果表明,玉米茎粗长轴和茎粗短轴与玉米鲜质量和干质量的相关性高于玉米株高,多元回归模型H+L+S、L×S和逐步回归模型具有较高的拟合精度,其对玉米鲜质量、干质量的决定系数高于0.874和0.877,均方根误差分别小于7.363和0.801 g,且单因素方差分析表明,3个模型之间没有明显差异,模型交叉验证结果表明,3个模型都具有较好的稳定性和预测能力。应用上述3个模型对玉米大喇叭口期的生物量进行预测,预测结果表明,模型对平展型玉米的预测精度优于紧凑型玉米。对平展型玉米生物量的预测中,逐步回归模型预测效果最优,其对玉米鲜质量、干质量的决定系数分别为0.866、0.875,均方根误差分别为30.790和2.752 g,相对均方根误差分别为13.53%、11.41%。该研究表明,利用玉米株高、茎粗长轴、茎粗短轴可实现对玉米生物量的估测,且对平展型玉米具有较好的预测效果。     

基于遗传算法的波长选择方法对土壤有机碳预测模型影响

以黑龙江农田黑土为研究对象,利用遗传算法(GA)波长选择结合偏最小二乘法(PLS)回归建立土壤有机碳(SOC)的预测模型。通过设定以下GA参数:波长选择数量上限k、初始种群大小P及迭代次数N,采用单点优化方式逐一确定各参数。结果表明,在主成份数为7的情况下,当GA的参数取N=300、P=300、k=50时,GA模型最优;模型的校正决定系数R2=0.922、校正均方根误差RMSEC=1.74、交叉检验均方根误差RMSECV=1.80;模型的预测决定系数R2=0.931、预测均方根误差RMSEP=1.84、预测相对误差RPD=3.81。与原始光谱的PLS模型相比,R2由0.900提升至0.922,RPD由3.38提升至3.81。结果表明,通过GA进行波长选择能够优化模型,提升模型稳定性以及预测精确性。     

苹果质地品质近红外无损检测和指纹分析

为了探索近红外光谱快速无损检测苹果质地品质的方法,采集240个苹果样本的近红外光谱( 波长 8002500 nm),通过解析光谱图和进行不同的预处理,利用偏最小二乘法(PLS)和多元线性回归(MLR)建立回归模型和确定特征指纹图谱.基于波长范围为1300～2500 nm,PLS结合多元散射校正(MSC)所建模型的预测效果最好,硬度模型的预测标准偏差(RMSEP)和决定系数(R2)分别为0.226 kg/cm2、96.52%,脆度模型的 RMSEP和R2分别为0.243 kg/cm2、97.15%.用权重法基于PLS模型选择的硬度特征波长为1657、1725、1790、2455、1929、2304 nm,脆度特征波长为1613、1725、1895、2304、2058、2087、2396 nm,经MLR模型检验,特征波长与苹果的硬度和脆度有很高的相关性,硬度的RMSEP和R2分别为0.271 kg/cm2、90.30%,脆度的RMSEP和R2分别为0.304kg/cn2、91.64%.结果表明,PLS模型和特征指纹光谱均能准确预测苹果的质地品质,为苹果质地品质的评价提供了快速、直观、简便、可行的新方法.     

基于高光谱图像的茶树LAI与氮含量反演

为了对茶树进行实时、快速、无损的叶面积指数LAI和氮含量检测,该文以英红九号茶树为试验对象,利用便携式高光谱成像仪采集光谱数据、人工破坏性采摘叶片进行叶面积指数的计算以及传统化学方法测量叶片氮含量,比较不同高光谱特征变换形式与LAI和氮含量之间的相关性,并选择其中相关系数较高的高光谱特征变量作为自变量,分别采用线性、指数、对数和抛物线表达式建立LAI和氮含量的回归模型。结果显示:在多种高光谱数据变量建立的模型中,以绿峰反射率R_g为自变量的对数拟合模型最佳,其拟合样本的决定系数R~2和验证样本的均方根误差RMSE值分别为0.9和0.087 6。以植被指数变量VI_4(红边面积/黄边面积)与氮含量建立的指数模型为最佳建模效果,拟合样本的决定系数R~2和验证样本的均方根误差RMSE值分别为0.830 3和0.102 9,研究结果可为茶树叶面积指数LAI和营养成分的无损检测提供参考。     

Nondestructive prediction of total phenolics, flavonoid contents, and antioxidant capacity of rice grain using near-infrared spectroscopy

Phytochemicals such as phenolics and flavonoids, which are present in rice grains, are associated with reduced risk of developing chronic diseases such as cardiovascular disease, type 2 diabetes, and some cancers. The phenolic and flavonoid compounds in rice grain also contribute to the antioxidant activity. Biofortification of rice grain by conventional breeding is a way to improve nutritional quality so as to combat nutritional deficiency. Since wet chemistry measurement of phenolic and flavonoid contents and antioxidant activity are time-consuming and expensive, a rapid and nondestructive predictive method based on near-infrared spectroscopy (NIRS) would be valuable to measure these nutritional quality parameters. In the present study, calibration models for measurement of phenolic and flavonoid contents and antioxidant capacity were developed using principal component analysis (PCA), partial least-squares regression (PLS), and modified partial least-squares regression (mPLS) methods with the spectra of the dehulled grain (brown rice). The results showed that NIRS could effectively predict the total phenolic contents and antioxidant capacity by PLS and mPLS methods. The standard errors of prediction (SEP) were 47.1 and 45.9 mg gallic acid equivalent (GAE) for phenolic content, and the coefficients of determination ( r (2)) were 0.849 and 0.864 by PLS and mPLS methods, respectively. Both PLS and mPLS methods gave similarly accurate performance for prediction of antioxidant capacity with SEP of 0.28 mM Trolox equivalent antioxidant capacity (TEAC) and r (2) of 0.82. However, the NIRS models were not successful for flavonoid content with the three methods ( r (2) < 0.4). The models reported here are usable for routine screening of a large number of samples in early generation screening in breeding programs.     

基于dbiPLS-SPA变量筛选的固态发酵湿度近红外光谱检测

为了提高基于近红外光谱技术的固态发酵关键过程参数——湿度快速检测的精度和稳定性,研究采用动态反向区间偏最小二乘（dbiPLS）法结合连续投影算法（SPA）进行最佳光谱子区间和特征组合变量的筛选,通过交互验证法确定偏最小二乘（PLS）模型的主成分因子数,并以预测均方根误差（RMSEP）和相关系数（Rp）作为模型的评价标准。试验结果显示,最佳dbiPLS-SPA模型筛选的组合变量个数为8,其RMSEP和Rp分别为1.1795%（质量分数）和0.9430。试验结果表明,dbiPLS-SPA是一个有效的波长组合变量筛选方法,可简化模型结构、增强模型精度和稳健性。     

Influence of the number of samples on prediction error of visible and near infrared spectroscopy of selected soil properties at the farm scale

This paper reports on the influence of the number of samples used for the development of farm‐scale calibration models for moisture content (MC), total nitrogen (TN) and organic carbon (OC) on the prediction error expressed as root mean square error of prediction (RMSEP) for visible and near infrared (vis‐NIR) spectroscopy. Fresh (wet) soil samples collected from four farms in the Czech Republic, Germany, Denmark and the UK were scanned with a fibre‐type vis‐NIR, AgroSpec spectrophotometer with a spectral range of 305–2200 nm. Spectra were divided into calibration (two thirds) and prediction (one third) sets and the calibration spectra were subjected to a partial least squares regression (PLSR) with leave‐one‐out cross‐validation using Unscrambler 7.8 software. The RMSEP values of models with a large sample number (46–84 samples from each farm) were compared with those of models developed with a small sample number (25 samples selected from the large sample set of each farm) for the same variation range. Both large‐set and small‐set models were validated by the same prediction set for each property. Further PLSR analysis was carried out on samples from the German farm, with different sample numbers of the calibration set of 25, 50, 75 and 100 samples. Results showed that the large‐size dataset models resulted in smaller RMSEP values than the small‐size dataset models for all the soil properties studied. The results also demonstrated that with the increase in sample number used in the calibration set, RMSEP decreased in almost linear fashion, although the largest decrease was between 25 and 50 samples. Therefore, it is recommended that the number of samples should be chosen according to the accuracy required, although 50 soil samples is considered appropriate in this study to establish calibration models of TN, OC and MC with smaller expected prediction errors as compared with smaller sample numbers.     

近红外光谱法测定玉米秸秆饲用品质

为了对玉米秸秆的饲用品质进行可靠、便捷、快速的分析和评价,该研究以不同品种、密度、氮肥和水分处理的不同发育时期和不同部位玉米秸秆为试验材料,应用近红外光谱（NIRS）技术和偏最小二乘法（PLS）,采用一阶导数+中心化+多元散射校正的光谱数据预处理方法,构建了玉米秸秆体外干物质消化率（IVDMD）、酸性洗涤纤维（ADF）、中性洗涤纤维（NDF） 和可溶性糖（WSC）含量的NIRS分析模型。所建立的IVDMD、ADF、NDF和WSC含量的NIRS校正模型决定系数（R2cal）分别为0.9906、0.9870、0.9931和0.9802,交叉验证决定系数（R2cv）分别为0.9593、0.9413 、0.9678和0.9342,外部验证决定系数（R2val）分别为0.9549、0.9353、0.9519和0.9191,各项标准差（SEC、SECV和SEP）为0.935～1.904,相对分析误差（RPD）均大于3。结果表明,各参数的NIRS分析模型可用于玉米秸秆饲用品质的分析和品种选育的快速鉴定。     

基于近红外光谱的沼液挥发性脂肪酸含量快速检测

挥发性脂肪酸（Volatile Fatty Acids，VFA）作为厌氧发酵过程的重要中间产物，其在厌氧反应器中的累积能够反映出产甲烷菌的不活跃状态或厌氧发酵条件的恶化。为了实现对农牧废弃物厌氧发酵进行过程分析和状态监控，将近红外光谱（Near Infrared Spectroscopy，NIRS）与偏最小二乘（Partial Least Squares，PLS）相结合构建玉米秸秆和畜禽粪便厌氧发酵液乙酸、丙酸和总酸含量快速检测模型。将竞争自适应重加权采样法（Competitive Adaptive Reweighted Sampling，CARS）与遗传模拟退火算法（Genetic Simulated Annealing algorithm，GSA）相结合构建CARS-GSA算法对沼液中的乙酸、丙酸和总酸进行特征波长优选，原始光谱数据1 557个波长点经预处理和波长优选后，得到乙酸、丙酸和总酸特征波长变量分别为135、101和245个，建立的回归模型验证决定系数分别为0.988、0.923和0.886，预测均方根误差（Root Mean Squared Error of Prediction，RMSEP）分别为0.111、0.120和0.727，相对分析误差分别为9.685、3.685和3.484，与全谱建模相比RMSEP分别减少了17.78%、15.49%和1.22%，能够满足农牧废弃物厌氧发酵过程发酵液中乙酸和丙酸含量的快速检测需求，基本满足总酸的检测需求。结果表明，通过构建CARS-GSA算法优选乙酸、丙酸和总酸的敏感波长变量，参与建模的波长点数量显著减少，有效降低了变量维度和模型复杂度，提升了回归模型检测精度和预测能力，为快速准确检测沼液VFA提供了新途径。     

(Methyl)Mercury,Arsenic, and Lead Contamination of the World’s Largest Wastewater Irrigation System: the Mezquital Valley (Hidalgo State—Mexico)

We investigate the potential of near-infrared (NIR) spectroscopy to predict some heavy metals content (Zn, Cu, Pb, Cr and Ni) in several soil types in Stara Zagora Region, South Bulgaria, as affected by the size of calibration set using partial least squares (PLS) regression models. A total of 124 soil samples from the 0–20 and 20–40 cm layers were collected from fields with different cropping systems. Total Zn, Cu, Pb, Cr and Ni concentrations were determined by Atomic Absorption Spectrometry. Spectra of air dried soil samples were obtained using an FT-NIR Spectrometer (spectral range 700–2,500 nm). PLS calibration models were developed with full-cross-validation using calibration sets of 90 %, 80 %, 70 % and 60 % of the 124 samples. These models were validated with the same prediction set of 12 samples. The validation of the NIR models showed Cu to be best predicted with NIR spectroscopy. Less accurate prediction was observed for Zn, Pb and Ni, which was classified as possible to distinguish between high and low concentrations and as approximate quantitative. The worst model performance in cross-validation and prediction was for Cr. Results also showed that values of root mean square error in cross-validation (RMSE_cv) increased with decreasing number of samples in calibration sets, which was particularly clear for Cu, Pb, Ni and Cr content. A similar tendency was observed in the prediction sets, where RMSEP values increased with a decrease in the number of samples, particularly for Pb, Ni and Cr content. This tendency was not clear for Zn, while even an increase in RMSEP for Cu with the sample size was observed. It can be concluded that NIR spectroscopy can be used to measure heavy metals in a sample set with different soil type, when sufficient number of soil samples (depending on variability) is used in the calibration set.     

Near-infrared spectroscopy analysis of seed coats of common beans ( Phaseolus vulgaris L.): a potential tool for breeding and quality evaluation

Near-infrared spectroscopy (NIRS) is a well-established technique for determining the components of foods. Sample preparation for NIRS is easy, making it suitable for breeding and/or quality evaluation, for which a large number of samples should be analyzed. We aimed to assess the feasibility of NIRS to estimate parameters that seem to influence consumers' perception of the seed coat of common beans: dietary fiber (DF), uronic acids (UA), ashes, calcium, and magnesium. We used reference methods to analyze ground seed coats of 90 common bean samples with a wide range of genetic variability and cultivated at many locations. We registered the NIR spectra on intact beans and ground seed coat samples. We derived partial least-squares (PLS) regression equations from a set of calibration samples and tested their predictive power in an external validation set. For intact beans, only RER values for ashes and calcium are good enough for very rough screening. For ground seed coat samples, the RPD and RER values for ashes (3.49 and 14.09, respectively) and calcium (3.57 and 12.70, respectively) are good enough for screening. RPD and RER values for DF (2.60 and 9.15, respectively) and RER values for magnesium (6.57) also enable rough screening. A poorer correlation was achieved for UA. We conclude that NIRS can help in common bean breeding research and quality evaluation.     

基于近红外光谱土壤水分检测模型的适应性

由于土壤水分的近红外光谱定量分析模型精度依赖于样品状态,故土壤水分定量分析模型的适应性极其重要。以湖北地区的3种土壤为研究对象,利用偏最小二乘法交叉验证建立了处理后样品下的土壤水分分析模型,模型预测值与标准值的决定系数R2为0.9946,交叉验证预测均方差为0.801%,模型预测决定系数R2为0.9919,预测均方差为0.912%;利用主成分分析了未处理土壤样品与处理土壤样品得分图的差异,结果表明定量分析模型对未处理样品的预测精度降低;采用斜率/截距的方法修正了12个未处理样品的模型预测值,预测平均绝对值误差从0.78%降低到0.38%,结果表明斜率/截距校正法能较好的提高近红外光谱土壤水分定量分析模型的适应性。     

油用牡丹单粒种子含油量NIRS模型的建立

为了建立油用牡丹单粒种子含油量的近红外测定模型,便于高含油量单株的选育,采用索氏抽提法测试了200份油用牡丹凤丹单粒种子的含油量,并应用近红外反射光谱技术(NIRS)采集了200份样品的光谱数据,通过偏最小二乘法(PLS)和主成分回归法(PCR)构建了油用牡丹单粒种子含油量的数学模型。结果表明,索氏抽提法中,均匀粉碎后的油用牡丹籽样品干燥烘焙条件为105℃ 2 h,牡丹籽抽提时间为20 h,测出的含油量变化范围在10%~28%之间,籽油含量基本符合正态分布。NIRS法构建的模型最佳参数为:采用PLS法,光程固定,一阶导数消除背景,数据平滑处理采用Norris derivative filter的方法,平滑参数选用5和3。内部交叉检验校正相关系数r₁为0.980 1、预测相关系数r₂为0.957 6、校正均方根误差(RMSEC)为0.463、预测均方根误差(RMSEP)为0.705。外部检验相关系数达 0.957 6, 平均误差小于3%。本试验所构建的牡丹单粒种子含油量的NIRS模型可靠,可以用于分析油用牡丹单粒种子的含油量。     

Development and Evaluation of a Near‐Infrared Instrument for Single‐Seed Compositional Measurement of Wheat Kernels

A single‐kernel, near‐infrared reflectance instrument was designed, built, and tested for its ability to measure composition and traits in wheat kernels. The major objective of the work was targeted at improving an existing design concept of an instrument used for larger seeds such as soybeans and corn but in this case designed for small seeds. Increases in throughput were sought by using a vacuum to convey seeds without compromising measurement accuracy. Instrument performance was evaluated by examining measurement accuracy of wheat kernel moisture, protein content, and kernel mass. Spectral measurements were obtained on individual wheat kernels as they were conveyed by air through an illuminated tube. Partial least squares (PLS) prediction models for these constituents were then developed and evaluated. PLS single‐kernel moisture predictions had a root mean square error of prediction (RMSEP) around 0.5% MC wet basis; protein prediction models had an RMSEP near 0.70%. Prediction of mass was not as good but still provided a reasonable estimate of single‐kernel mass, with RMSEP values of 2.8–4 mg. Data showed that kernel mass and protein content were not correlated, in contrast to some previous research. Overall, results showed the instrument performed comparably to other single‐seed instruments or methods based on accuracy but with an increased throughput at a rate of at least 4 seeds/s.