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。     

近红外技术快速测定肉鸡粪便主要肥料成分含量的研究

该文探讨了利用近红外光谱分析技术(NIRS)快速测定肉鸡粪便主要肥料成分含量的可行性。在饲养试验过程中采集了肉鸡粪便样品183个，利用常规实验室分析方法测定了其中的总氮(TN)、总磷(TP)、总钾(TK)、铵态氮(AN)、有效磷(EP)含量，利用近红外光谱仪取得了样品在1421.5～2572.2 nm波段的光谱，并使用偏最小二乘法(PLS)结合交互验证(CV)和留一检验法(LOO)建立了肉鸡粪便肥料成分的近红外光谱定标模型，同时利用该模型对20个样品的总氮、总磷、总钾、铵态氮、有效磷的含量进行预测，得出的5种主要肥料成分的预测值和真实值(实验室经典化学分析方法测定值)之间具有显著的相关性，其相关系数分别为0.9574，0.9590，0.9870，0.9572和0.9650。预测标准差分别为0.0014，0.0012，0.0012，2.3041×10^-4和0.3814。结果表明，利用近红外光谱法对肉鸡粪便风干样品的主要肥料成分进行快速测定是可行的。     

黄绵土钾含量高光谱估算模型研究

为了研究可见/近红外光谱法估算渭北旱塬区黄绵土钾含量的可行性,以陕西省乾县试验田采集的120个土壤样品为研究对象,在室内进行土壤全钾、速效钾含量及反射光谱数据测量的基础上,应用多元线性回归(MLR)和偏最小二乘回归(PLSR)方法建立土壤钾含量的估算模型,并用独立样本进行验证。结果表明,以土壤光谱反射率一阶微分(DSSR)为自变量建立的多元线性回归模型(MLR)能进行土壤全钾含量准确估算。以波段深度一阶微分(DBD)为自变量建立的PLSR模型,验证集的决定系数(R2pre)大于0.90,预测均方根误差(RMSEpre)等于0.054,预测相对分析误差(RPDpre)等于3.310,是估算土壤全钾含量的最优模型;而以DSSR为自变量建立的PLSR模型,RPDpre值为1.619和1.572,是估算土壤速效钾含量的最优模型。本研究表明可见/近红外光谱结合多元线性回归和偏最小二乘回归方法能对渭北旱塬区黄绵土全钾含量进行快速、准确估算,但对速效钾含量仅能进行粗略估算。     

曲靖烤烟钾含量特征及其与主要生态因子关系研究

为了研究曲靖烤烟钾含量特征及其与主要生态因子的关系,在曲靖烟区采集烤烟样品3 506个,测定其钾含量,并采用决策树分析、回归分析等方法探究曲靖烤烟钾含量特征及其主要影响因素。结果表明,曲靖烟区烤烟钾含量平均为1.65%,且变幅较大(0.35%~3.60%),有81.86%的烟叶钾含量2.0%,钾含量在县域间差异达极显著水平;此外,海拔高度、土壤质地、土壤类型、土壤有机质、土壤速效钾、土壤有效钙、土壤有效镁均影响烤烟钾含量。海拔高度对烤烟钾含量影响的拐点分别为1 689、1 814和2 023m,且随海拔上升,烤烟钾含量下降;土壤质地粘性增加,烤烟钾含量随之增加;烤烟钾含量在不同土壤类型间差异达极显著水平,其中黄壤中的含量最高(2.07%),紫色土最低(1.59%);土壤有机质对烤烟钾含量影响的拐点分别为23.768和39.060 g·kg~(-1),且随土壤有机质增加,烤烟钾含量表现为逐渐增加趋势;土壤速效钾对烤烟钾含量影响的拐点分别为153.746、205.000和328.169 mg·kg~(-1),随土壤速效钾增加,烤烟钾含量表现为逐渐增加趋势;随土壤有效钙增加,烤烟钾含量呈下降趋势;随土壤有效镁增加,烤烟钾含量呈先上升高后下降的趋势,有效镁含量为300 mg·kg~(-1)左右时,烤烟钾含量达峰值;随着土壤钙钾比和镁钾比升高,烤烟钾含量呈减速下降的趋势。本研究结果对提高曲靖烟区烤烟钾含量具有一定指导意义。     

近红外光谱法预测羊肉化学成分的研究

对3个品种、3个部位的106个羊肉样品进行近红外光谱扫描,并测定其蛋白质、水分、脂肪含量,采用Unscrambler软件建立基于偏最小二乘法的近红外光谱预测模型。结果显示:样品水分含量近红外光谱校正决定系数为0.94,验证决定系数是0.86;蛋白质含量近红外光谱预测模型的校正决定系数为0.90,验证决定系数为0.72;脂肪含量近红外光谱校正决定系数0.81,验证决定系数0.64,由此可知近红外光谱用于羊肉品质检测具有可行性。本研究为羊肉化学成分的快速检测提供了基础。     

苏北滨海土壤碳酸钙含量反射光谱估算研究

土壤属性的快速、精确测定是实现现代精细农业的基础。本研究分析了江苏省北部滨海土壤的属性特征以及碳酸钙的可见-近红外反射光谱特征,探讨利用可见-近红外光谱估算滨海土壤碳酸钙含量的可行性,比较不同光谱反射率数据集、不同预处理方法以及不同建模方法定量反演的优劣。结果表明:(1)苏北滨海土壤有机质含量较低、碳酸钙含量较高,其光谱曲线在2 340 nm处有较明显的碳酸钙吸收特征;(2)滨海土壤碳酸钙含量与土壤的可见-近红外波段反射率呈正相关,且碳酸钙含量高低对于土壤的近红外波段反射率的影响高于可见光波段;(3)可见-近红外反射光谱可用于估算滨海土壤碳酸钙含量。就建模结果而言,381~2 459 nm波段反射光谱数据集、log(1/R)预处理、偏最小二乘回归三者结合的效果比较理想。     

紫花苜蓿幼苗耐盐性的近红外光谱鉴定

紫花苜蓿幼苗耐盐性快速鉴定对于耐盐种质资源筛选和耐盐新品种选育具有重要意义。脯氨酸和丙二醛是表征植物耐盐性的两种重要生化指标。研究应用便携式近红外仪和近红外光谱分析技术,结合偏最小二乘回归法,研究了40个不同紫花苜蓿品种幼苗耐盐性的脯氨酸和丙二醛含量,建立了新鲜样品和干燥样品的近红外漫反射光谱定量分析模型。研究结果表明：脯氨酸、丙二醛的近红外漫反射光谱分析效果均较好,校正模型决定系数R2 和验证集样品预测值与常规分析测定值的验证决定系数r2都大于0.85,两种样品脯氨酸的相对分析误差RPD值分别为1.72     

近红外反射技术建立合肥地区精米直链淀粉含量测定模型

以合肥地区种植的203份水稻材料为检测对象,用近红外反射技术采集光谱,常规化学方法测定精米直链淀粉含量。结果表明,定标样品的直链淀粉含量分布范围为3.439%～28.046%,代表性和连续性良好。采用多种计量数学处理方法和偏最小二乘法(PLS),优化建立了精米直链淀粉含量的定量分析预测模型。定标集(C-Set)样品数132个,相关系数(Rc)0.9278,定标标准差(SEC)1.6582;验证集(V-Set)样品数67个,相关系数(Rv)0.8736,预测标准差(SEP)1.9083,并证实所建立的模型在测定精米直链淀粉含量上具有很好的准确性和实用性,对合肥地区水稻品质育种及种质资源相关研究具有实用价值。     

基于可见-近红外反射光谱的土壤碳酸钙含量与反演效果关系研究

土壤可见-近红外反射光谱中包含了大量的土壤属性信息,研究人员根据土壤属性信息在光谱上的特征,对土壤属性进行定量反演。是否属性值越高,反演精度越高?目前对于属性含量与反演效果的定量关系尚不清楚。采集了我国西北地区黑河流域69个代表性干旱土剖面(292个发生层土样),以气量法测定其碳酸钙含量,使用Cary 5000分光光度计测定其可见-近红外光谱反射率,以样本量和离散度(变异系数)作为数据集划分标准,分别建立了11个相同样本量子集(A)和5个相近离散度子集(B),应用偏最小二乘回归(PLSR)算法对各子集进行土壤碳酸钙含量反演,以此探究碳酸钙含量与反演效果的定量关系。结果表明,碳酸钙可增加可见-近红外波段的光谱反射率,但利用可见近红外光谱反演土壤碳酸钙含量,其反演效果与碳酸钙含量关系不显著。     

羊肉色泽傅立叶变换近红外光谱定量分析方法研究

以从北京市、山西大同市、宁夏吴忠市3个地区筛选的有代表性的227份羊肉样品为试材,应用傅里叶变换近红外光谱技术探讨羊肉色泽无损检测的方法。以决定系数(R2)、校正标准差(RMSECV)和预测标准差(RMSEP)为近红外光谱检测模型的评价指标,采用偏最小二乘法(PLS)对近红外光谱信息与样品的色差e值进行拟合,确定最佳的光谱预处理方法、主成分数和光谱区间范围。结果表明:所选227个羊肉样品的色差e值分布范围为1.556～9.879,其中80%以上的样品e值在1～5之间,具有显著的代表性;在11995.5～4597.6cm-1的波段范围内,最佳主成分数为6时,近红外光谱经最大最小归一法处理后,建立的羊肉色泽预测模型精度最高,R2达到0.776,RMSECV为0.451;用此模型对预测集48个样品进行预测,预测值与实测值的相关系数(R)为0.835,RMSEP为0.517,该研究表明利用近红外光谱技术检测羊肉色泽可行。     

Near-Infrared Reflectance Spectroscopy as a Method for Determining Organic Carbon Concentrations in Soil

Near-infrared reflectance spectroscopy (NIRS) has the potential to be a reliable method for accurately quantifying soil organic carbon (SOC). The objective of this study was to evaluate NIRS as a method for predicting SOC. Partial least squares (PLS) regression was used to predict SOC from soil reflectance values or the first derivative of the reflectance values. Two model validation techniques were evaluated: One was a full cross-validation and in the other 30 percent of the samples were removed from the calibration data set and then tested using the calibrated model. Significant relationships were observed for predicted SOC when compared to laboratory-measured SOC for all models evaluated, regardless of validation technique. The prediction models using the first derivative of the reflectance values outperformed prediction models using the reflectance values alone. In conclusion, NIRS can be used as a quick and accurate method for measuring SOC.     

大米直链淀粉含量的近红外光谱分析

大米的直链淀粉含量是影响大米蒸煮和加工特性的最重要因素之一,常被用作蒸煮米质构特性评价指标。该文对不同粒度、不同类型大米样品进行了近红外光谱分析,建立了大米直链淀粉含量的预测模型,（精米样品）预测值与化学分析值的相关系数达0.95。预测标准差、平均相对误差分别为0.56和3.1%。     

Ash and Mineral Contents in Leaves of Woody Species: Analysis by Near‐Infrared Reflectance Spectroscopy

Abstract

Near‐infrared reflectance spectroscopy (NIRS) was evaluated for its effectiveness to determine ash and mineral concentrations [potassium (K), magnesium (Mg), copper (Cu), iron (Fe), and zinc (Zn)], in a total of 182 leaf samples of 17 woody species located in the central‐western region of the Iberian Peninsula. Chemical analysis revealed great variability in all leaf mineral elements. This variability was mainly related to differences in leaf habit (deciduous versus evergreen) and to differences in mean leaf longevity and among leaf age classes within evergreen species. A set of samples including all 17 species and leaf age classes was used to develop the calibration equations using multiple linear regression (MLR) and partial‐least squares regression (PLSR). The set of samples that did not enter in the calibration was used for external validation. In general, the most satisfactory results were obtained using PLSR and derivative transformations. Despite the strong heterogeneity of the samples included in the study, the results showed that NIRS can be employed as an effective tool, alternative to the more time‐consuming standard methods. The best predictive model was obtained for ash content. Models with acceptable accuracy were obtained in the prediction of K and Mg contents. However, their applicability for the determination of trace elements was more limited.     

Quality evaluation of sugar beet (Beta vulgaris) by near-infrared spectroscopy

The legal method (polarimetric measurement) for the determination of sucrose content and the wet chemical analysis for the quality control of sugar beet uses lead acetate. Because heavy metals are pollutants, the law could forbid their use in the future. Therefore, near-infrared spectroscopy (NIRS) was evaluated as a procedure to replace these methods. However, there are alternatives to lead clarification, such as the use of aluminum salts, which have been applied at many sugar companies. The real advantage of NIRS is in speed and ease of analysis. The aim of this study was to determine simultaneously the concentration of several components which define the industrial quality of beets. The first objective was the determination of sucrose content, which determines the sugar beet price. The standard error of prediction (SEP) was low: 0.11 g of sucrose/100 g of fresh beet. NIRS was also able to determine other beet quality parameters: brix, marc, glucose, nitrogen, sodium, potassium, sugar in molasses (i.e. sucrose in molasses), and juice purity. The results concerning brix, marc, sugar in molasses, and juice purity were satisfactory. NIRS accuracy was lower for the other parameters. Nevertheless, RPD (ratio standard deviation of concentration/SEP) and RER (ratio concentration range/SEP ratio) show that NIRS might be used for the sample screening on nitrogen, potassium, sodium, and glucose content.     

NIRS分析技术在农业中的应用进展

近红外光谱分析技术是一种间接测量技术。它是应用化学计量学方法建立校正模型,从而实现对未知样品的定性或者定量分析,已经在很多领域得到应用。该文论述了近红外光谱分析技术的分析步骤及其技术特点,以及近年来在国内农产品品质分析、食品分析、饲料工业分析、土壤分析、农产品在线快速检测分析等农业领域中的应用研究现状,并分析该技术在应用中存在的主要问题和相应的解决方案。同时指出了未来几年国内关于近红外光谱仪器硬件的开发及其化学计量学方法和模型优化方面的进一步探索,将成为国内未来几年近红外光谱技术研究的热点。     

Application of Near Infrared Reflectance Spectroscopy for the Assessment of Soil Quality in a Long‐Term Pasture

Abstract

The objective of the present study was to assess the ability of near infrared reflectance spectroscopy (NIRS) to analyze chemical soil properties and to evaluate the effects of different phosphorus (P) and potassium (K) fertilization rates on soil quality in different layers of a long‐term pasture. The NIRS calibrations were developed for humus, total Kjeldahl nitrogen (N_Kjeldahl), and several humic substances (HA1, “mobile” humic acids fraction; ΣHA, sum of humic acids; FA1, “mobile” fulvic acids; ΣFA, sum of fulvic acids, etc.) using soil samples of rather heterogeneous origin, collected during 1999–2003. Different spectral preprocessing and the modified partial least squares (MPLS) regression method were explored to enhance the relation between the spectra and measured soil properties. The equations were employed for the quality prediction of a sod gleyic light loam (Cambisol) in five PK fertilization treatments. The soil was sampled in 2000 and 2003 in three field replicates at depths of 0–10, 10–20, 20–30, and 30–50 cm, n=60 samples yr^?1. The best coefficients of correlation, R², between the reference and NIRS‐predicted data were as follows: for N_Kjeldahl, 0.965; humus, 0.938; HA1, 0.903; HA2, 0.905; HA3, 0.924; ΣHA, 0.904; and FA1, 0.911; and ΣFA, 0.885. Our findings suggest that it is feasible to use NIRS for the assessment of the effects of the inorganic PK fertilizer on the soil quality in different depths of a long‐term pasture.     

Quantification of exopolysaccharide, lactic acid, and lactose concentrations in culture broth by near-infrared spectroscopy

Near-infrared spectroscopy (NIRS) was used for the simultaneous prediction of exopolysaccharide (EPS; 0-3 g/L) and lactic acid (0-59 g/L) productions as well as lactose (0-68 g/L) concentration in supernatant samples from pH-controlled batch cultures of Lactobacillus rhamnosus RW-9595M in supplemented whey permeate medium. To develop calibration equations, the correlation between the second derivative of 164 NIRS transmittance spectra and concentration data obtained with reference methods was calculated at the wavelength between 1653-1770 and 2041-2353 nm, using a partial least-squares method (PLS). The lactic acid and lactose concentrations were measured by HPLC, and the EPS concentration was estimated by a new ultrafiltration method. The PLS correlation coefficient (R(2)) and the standard error of cross-validation for the calibrations were 91% and 0.26 g/L for EPS, 99% and 2.54 g/L for lactic acid, and 98% and 3.32 g/L for lactose, respectively. The calibration equations were validated with 45 randomly selected culture samples from 6 cultures that were not used for calibration. A high agreement between data of the reference methods and those of NIRS was observed, with correlation coefficients and standard errors of prediction of 99% and 1.64 g/L for lactic acid, 99% and 4.5 g/L for lactose, and 91% and 0.32 g/L for EPS. The results suggest that NIRS could be a useful method for rapid monitoring and control of EPS lactic fermentations.     

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。     

应用田间行走式红外光谱进行土壤碳含量估测研究

对应用田间行走式设备获取的土壤红外光谱数据,通过特征变换和特征选择相结合,以提高所建立土壤碳校正模型的预测精度。首先应用独立成分分析（ICA）、主成分分析(PCA)和小波分析(WA)对土壤红外光谱数据进行特征变换,然后分别应用无信息变量消除法(UVE)、连续投影算法(SPA)、无信息变量消除结合连续投影算法(UVE-SPA)、基于遗传算法和偏最小二乘法的变量选择法(GA-PLS)来进行特征选择,基于所选择的特征建立了土壤碳校正模型。结果表明,通过ICA进行特征变换,然后进行特征选择,可以建立比直接对光谱数据进行波长选择精度更好的预测模型;而WA或PCA与特征选择方法结合,只能获得与对光谱数据直接进行波长选择相近的效果。因此,针对田间条件下通过行走式设备获得的光谱数据由于受复杂的环境条件下干扰多的情况,可以将ICA与特征选择方法结合起来对光谱数据进行特征变换和选择,以建立更可靠的土壤碳含量预测模型。