Prediction of pork quality with near infrared spectroscopy (NIRS): 1. Feasibility and robustness of NIRS measurements at laboratory scale

The objective was to study prediction of pork quality by near infrared spectroscopy (NIRS) technology in the laboratory. A total of 131 commercial pork loin samples were measured with NIRS. Predictive equations were developed for drip loss %, colour L*, a*, b* and pH ultimate (pHu). Equations with R(2)>0.70 and residual prediction deviation (RPD)≥1.9 were considered as applicable to predict pork quality. For drip loss% the prediction equation was developed (R(2) 0.73, RPD 1.9) and 76% of those grouped superior and inferior samples were predicted within the groups. For colour L*, test-set samples were predicted with R(2) 0.75, RPD 2.0, colour a* R(2) 0.51, RPD 1.4, colour b* R(2) 0.55, RPD 1.5 and pHu R(2) 0.36, RPD 1.3. It is concluded that NIRS prediction equations could be developed to predict drip loss% and L*, of pork samples. NIRS equations for colour a*, b* and pHu were not applicable for the prediction of pork quality on commercially slaughtered pigs.     

Application of NIRS for predicting fatty acids in intramuscular fat of rabbit

The aim of this study was to evaluate the use of near infrared reflectance spectroscopy (NIRS) for predicting fatty acid content in intramuscular fat to be applied in rabbit selection programs. One hundred and forty three freeze-dried Longissimus muscles (LM) were scanned by NIRS (1100-2498nm). Modified Partial Least Squares models were obtained. Equations were selected according to standard error of cross validation (SECV) and coefficient of determination of cross validation (R(2)(CV)). Residual predictive deviation of cross validation (RPD(CV)) was also studied. Accurate predictions were reported for IMF (R(2)(CV)=0.98; RPD(CV)=7.57), saturated (R(2)(CV)=0.96; RPD(CV)=5.08) and monounsaturated FA content (R(2)(CV)=0.98; RPD(CV)=6.68). Lower accuracy was obtained for polyunsaturated FA content (R(2)(CV)=0.83; RPD(CV)=2.40). Several individual FA were accurately predicted such as C14:0, C15:0, C16:0, C16:1, C17:0, C18:0, C18:1 n-9, C18:2 n-6 and C18:3 n-3 (R(2)(CV)=0.91-0.97; RPD(CV)>3). Long chain polyunsaturated FA and C18:1 n-7 presented less accurate prediction equations (R(2)(CV)=0.12-0.82; RPD(CV)<3).     

Perspective of NIRS measurements early post mortem for prediction of pork quality

The potential of near-infrared spectroscopy (NIRS) measurements early post mortem was investigated to predict ultimate drip loss, colour, tenderness and intra-muscular fat of pork. Three locations (M. longissimus thoracis, M. longissimus lumborum and M. semimembranosus) in 102 pig carcasses were tested at the end of the slaughter line. A priori variation in pork quality was introduced using an experimental design covering: genotype, lairage time, pre-slaughter handling and day of slaughter. At 1 h post mortem a diode array VIS/NIR instrument (Zeiss MCS 511/522, 380-1700 nm) equipped with a surface fibre optic probe was used and at 1 day post mortem ultimate pH, drip loss, colour and shear force was measured on similar locations. Results indicated that it was possible to predict intra-muscular fat content (correlation (R(2) of 0.35 with multiple linear regression), standard error of prediction (SEP)=3.6 g/kg), but the configuration has to be refined for on-line application (bigger aperture). For drip loss no correlation was achieved with the PLS method. Even extremes (low drip loss (<2.5%) or high drip loss (>4.5%)) in drip loss were not discriminated. Predicting drip loss with NIRS early post mortem is not successful, although NIRS in the slaughter line has potential as a fast predictor of intra-muscular fat. Possibilities for using the NIRS technique to get to know more about muscle metabolism and post mortem changes are promising.     

基于不同预处理高光谱信息的鸡肉滴水损失率快速预测研究

本文旨在通过挖掘不同预处理高光谱(900~1700 nm)信息构建鸡肉滴水损失率的快速预测模型。首先采集每个鸡肉样本高光谱图像并提取图像感兴趣区域内的平均光谱信息,经基线校正(BC)、标准正态变量校正(SNV)、多元散射校正(MSC)、高斯滤波平滑(GFS)、归一化校正(NC)等五种光谱不同预处理,利用偏最小二乘回归(Partial Least Squares Regression,PLSR)算法构建光谱信息与鸡肉滴水损失率之间的定量关系。然后分别基于回归系数法(Regression Coefficient,RC)、连续投影算法(Successive Projections Algorithm,SPA)和逐步回归算法(Stepwise)筛选出对模型精度影响较大的最优波长优化全波段PLS模型。结果显示,基于BC光谱的全波段PLSR模型(BC-PLSR)预测鸡肉滴水损失率效果更好(r_P=0.95,RMSEP=0.29%,RPD=3.07,ΔE=0.0024%)。利用Stepwise法从BC光谱中选取的14个最优波长(900.6、903.8、905.5、907.1、917.0、997.7、1162.2、1272.4、1354.8、1369.6、1410.8、1425.6、1584.1和1695.1 nm)建立的SW-BC-PLSR模型(r_P=0.97,RMSEP=0.24%,RPD=3.82,ΔE=0.0012%)和多元线性回归(Multiple Linear Regression,MLR)模型SW-BC-MLR(r_P=0.97、RMSEP=0.22%、RPD=4.19,ΔE=0.0036%)预测鸡肉滴水损失率效果均良好。本试验表明,基于近红外高光谱信息可潜在实现鸡肉滴水损失率的快速预测。     

Prediction of pork quality attributes from near infrared reflectance spectra

Near infrared spectroscopy (NIRS) is one of the most promising techniques for large-scale meat quality evaluation. We investigated the potential of NIRS-based models to predict drip loss and shear force of pork samples. Near infrared reflectance spectra (1000-2500 nm), water-holding capacity, shear force, ultimate pH, and colour (L(?), a(?), b(?)-value) of 96 pork longissimus muscles were recorded at 2 days post mortem. Stepwise multiple linear regression (SMLR) and partial least squares regression (PLSR) analyses were used to formulate models for drip loss and shear force. Prediction models for drip loss correlated moderately strong with measured drip loss (R=0.71-0.74), which is similar to the correlation obtained using a combination of ultimate pH, filter paper test, and L(?)-value (R=0.74). The current results indicate that NIRS enables the classification of pork longissimus muscles with a superior or inferior water-holding capacity as having a drip loss lower than 5% or higher than 7%. No useful models could be constructed for shear force.     

Relationships between ultimate pH and microbial, chemical, and physical characteristics of vacuum-packaged pork loins

ABSTRACT:  This study evaluated the effect of ultimate pH (pHu) of pork on shelf life based upon microbial growth, drip loss, and oxidative rancidity (2-thiobarbituric acid [TBA] procedure) in vacuum-packaged loins stored at 4 °C. Glucose and lactate concentrations of the pork loins were also measured. Thirty-six pork loins (pH = 5.56 to 6.57) were collected at a commercial slaughter facility 1-d postslaughter. All pigs were from the same genetic line. Loins were grouped by pH (group: pH range): A: 5.55 to 5.70, B: 5.71 to 5.85, C: 5.86 to 6.00, D: 6.01 to 6.15, and E: > 6.16. They were analyzed at days 0, 6, 14, 24, and 34. For aerobic plate counts, groups A and B were significantly lower than C through E, while psychrotrophic or Enterobacteriaceae counts of groups A and A through C were significantly lower than groups B through E and D and E, respectively. Lactic acid bacteria counts were not significantly influenced by pHu. Group A had higher glucose concentrations than groups C through E and higher lactate concentrations than groups D through E on most sampling days. Group A had a higher TBA value than group E at days 0 and 34. Group A displayed greater drip loss than groups D and E at day 6 and groups B through D on days 24 and 34. Based on the microbial and drip loss results, a pork loin pHu of 5.8 to 5.9 appears to be optimum to provide a vacuum-packaged shelf life of at least 24 d with minimum drip loss.     

Myosin heavy chain isoforms account for variation in pork quality

The objective of this study was to determine the impact of myosin heavy chain (MyHC) isoforms (I, IIB, IIA and IIX) on pork quality traits of halothane (HAL)-negative (NN) and halothane-carrier (Nn) pigs. Gilts (n=32) were assigned to a 2×2 factorial of genetic population (GP) and slaughter weight (WT; 120 and 135 kg). Classical meat quality characteristics were collected and MyHC content was determined on muscle samples. Regression equations for pork quality and carcass composition traits were determined. Only I/IIB accounted for variation in drip loss of NN gilts (R(2)=0.18), while GP related to drip loss in Nn gilts (R(2)=0.70). Type I MyHC content explained variation in ultimate (24 h) muscle pH of NN gilts (R(2)=0.09), while I/IIB, I/IIX and IIB/IIX were significant for Nn gilts (R(2)=0.99). I/IIA, I/IIX, IIB/IIX and GP accounted for variation in Hunter Color a (redness) values of NN gilts (R(2)=0.69), while IIB, IIA, IIB/IIA and GP were significant for Nn gilts (R(2)=0.97). Overall, fiber type composition accounts for a larger proportion of variation in the quality traits of Nn compared to NN gilts.     

Evaluation of the quality of deep frying oils with Fourier transform near-infrared and mid-infrared spectroscopy

Deep frying oils are subjected to high temperature and prolonged heating that may lead to a series of quality and safety problems for fried foods. This study evaluated the quality of deep frying oils collected from a local college canteen (n = 132) with Fourier transform mid-infrared (FT-IR) and Fourier transform near-infrared (FT-NIR) spectroscopy. Partial least squares (PLS) regression was used to correlate spectral data with free fatty acids (FFA) and peroxide (PO) values of frying oils. The coefficient of determination (R(2)), standard error of prediction (SEP), and the RPD (ratio of the standard deviation of data set to the SEP) were used as indicators for the predictability of the PLS models. The FT-IR and FT-NIR methods exhibited similar predictability for the FFA values (FT-IR: R(2) = 0.954, SEP = 0.14, RPD = 4.48; FT-NIR: R(2) = 0.948, SEP = 0.14, RPD = 4.38). Although the predictability of the FT-IR method for the PO values was not as satisfactory as that of the FT-NIR method (FT-IR: R(2) = 0.893, SEP = 6.17, RPD = 2.93; FT-NIR: R(2) = 0.953, SEP = 4.15, RPD = 4.36), both FT-IR and FT-NIR methods could be used as simple and rapid approaches to determining the quality of deep frying oils.     

Effects of fasting prior to slaughter on technological and sensory properties of the loin muscle (M. longissimus dorsi) of pigs

The aim of this study was to investigate the effect of pre-slaughter fasting time, sex and feeding regime on water-holding capacity (WHC), colour and sensory properties and their relationship with pH in M. longissimusdorsi (LD) in pigs. 270 pigs of the commercial Norwegian crossbreed Noroc (Norwegian Landrace × Yorkshire sow and Norwegian Landrace × Duroc boar) were used involving two sexes (gilts and castrates), two feeding regimes (restricted and ad libitum) and four fasting treatments: (F4) 4 h fasting (control), (F175) 17.5 h fasting on the farm, (FO175) 17.5 h fasting overnight at the abattoir, and (FO265) 26.5 h fasting overnight at the abattoir. Additionally, the pigs experienced two abattoir lairage times as fasting treatments F4 and F175 had a lairage time of 1.5 h, while fasting treatments FO175 and FO265 had a lairage time of 23.0 h. A short fasting time of 4 h led to a delayed decline in pH post-mortem and a lower ultimate pH (pHu) in the LD compared with a fasting time of 26.5 h which resulted in a rapid pH decline early post-mortem and a high pHu. Prolonged fasting reduced drip loss, resulted in a darker colour and tended to improve tenderness of the LD. Castrates showed lower drip loss, higher lightness and improved tenderness and juiciness compared with gilts, while ad libitum feeding improved tenderness compared to restricted feeding. There are obvious negative relationships between pHu and drip loss, lightness and tenderness of LD.     

Prediction of fatty acid content in rabbit meat and discrimination between conventional and organic production systems by NIRS methodology

To investigate the feasibility of using the NIRS methodology to analyse the fatty acid content of rabbit meat and to discriminate between conventional and organic production, the meat of a hind leg of 119 rabbits was scanned between 1100 and 2498 nm and 104 samples were sent to the laboratory for reference analysis of fatty acids by gas chromatography. A commercial spectral analysis program (WINISI-2, v. 1.04) was used to process the data and to develop chemometric models. The better calibration equation for each fatty acid, leading to a higher determination coefficient of cross-validation (r²) and low standard error of cross-validation (SECV) was retained. Prediction of linoleic, palmitic, palmitoleic and oleic acid content was excellent or good (r² between 0.90 and 0.70); prediction of arachidonic, stearic, α-linolenic and eicosatrienoic FA has r² between 0.69 and 0.50. However, miristic, vaccenic, icosaenoic and eicosadienoic FA are problematic to predict. When fatty acids were grouped, the r² of the calibration equations were: 0.85 for saturated FA, 0.83 for MUFA, 0.92 for PUFA and 0.91 for n − 6 FA, indicating excellent or good prediction. Prediction of α-linolenic FA (r² = 0.59) needs more precision. The obtained equations have been applied for predicting meat fatty acid composition of both groups of production systems, conventional and organic, for an other 52 rabbit meat samples (2 × 26). Meat of the organic source had lower (p = 0.000) monounsaturated FA (30.54% vs. 34.64%) and higher (p = 0.019) polyunsaturated FA (27.28% vs. 23.66%) than rabbit meat from the conventional system, while the saturated FA content was similar (42%) in both groups. The discriminant model correctly classified (98%) between conventional or organic produced rabbit meat.     

Bayesian meta-analysis of the effect of fasting, transport and lairage times on four attributes of pork meat quality

Technological meat quality is a significant economic factor in pork production, and numerous publications have shown that it is strongly influenced both by genetic status and by rearing and slaughter conditions. The quality of meat is often described by meat pH at different times postmortem, as well as by color and drip loss. A meta-analysis based on a database built from 27 studies corresponding to a total of 6526 animals classified was carried out. The purpose of this meta-analysis was to study the effect of fasting, lairage and transport durations on four main attributes of the technological pork meat quality. A Bayesian hierarchical meta-regression approach was adopted.The results of our meta-analysis showed that fasting time had a significant effect on pH measured 24 h post-mortem (pHu) and drip loss (DL) measured in longissimus muscle. While, lairage affected only the pHu in semimembranosus muscle. Interestingly, we found that DL was the lone attribute that was affected by transport time and its interaction with fasting time.     

Technical note: Near infrared reflectance spectroscopy to predict fecal indigestible neutral detergent fiber for dairy cows

In vitro and in situ procedures performed to estimate indigestible neutral detergent fiber (iNDF) in forage or fecal samples are time consuming, costly, and limited by intrinsic factors. In contrast, near infrared reflectance spectroscopy (NIRS) has become widely recognized as a valuable tool for accurately determining chemical composition and digestibility parameters of forages. The aim of this study was to build NIRS calibrations and equations for fecal iNDF. In total, 1,281 fecal samples were collected to build a calibration data set, but only 301 were used to develop equations. Once dried, samples were ground and chemically analyzed for crude protein, ash, amylase and sodium sulfite–treated NDF corrected for ash residue (aNDFom), acid detergent fiber, acid detergent lignin, and in vitro digestion at 240 h to estimate iNDF (uNDF240). Each fecal sample was scanned using a NIRSystem 6500 instrument (Perstorp Analytical Inc., Silver Spring, MD). Spectra selection was performed, resulting in 301 sample spectra used to develop regression equations with good accuracy and low standard error of prediction. The standard error of calibration (SEC), cross validation (SECV), and coefficients of determination for calibration (R²) and for cross validation (1 ? VR, where VR = variance ratio) were used to evaluate calibration and validation results. Moreover, the ratio performance deviation (RPD) and ratio of the range of the original data to SECV (range/SECV; range error ratio, RER) were also used to evaluate calibration and equation performance. Calibration data obtained on fiber fractions aNDFom (R² = 0.92, 1 ? VR = 0.87, SEC = 1.48, SECV = 1.89, RPD = 2.80, and RER = 20.19), uNDF240 (R² = 0.92, 1 ? VR = 0.86, SEC = 1.65, SECV = 2.24, RPD = 2.57, and RER = 14.30), and in vitro rumen aNDFom digestibility at 240 h (R² = 0.90, 1 ? VR = 0.85, SEC = 2.68, SECV = 3.43, RPD = 2.53, and RER = 14.0) indicated the predictive equations had good predictive value.     

枣醋发酵液中pH值与总酸含量的高光谱图像技术定量分析

利用高光谱图像技术（HS-IT）对灵武枣醋发酵过程中pH值和总酸含量进行定量分析,并通过偏最小二乘法（PLS）建立定量分析模型,同时采用竞争性自适应加权算法（CARS）和遗传算法（GA）对整个谱区进行特征波长筛选。以决定系数（R2）、预测均方根偏差（RMSEP）、相对分析误差（RPD）以及最佳主因子数作为模型质量的评价参数,其中使用CARS进行的波长筛选法对模型的优化效果最佳,pH值和总酸含量的R2分别达到0.928 4和0.935 1,RMSEP分别为0.122 6和0.301 5,RPD分别为3.75和3.91。结果表明,CARS-PLS法可提高枣醋发酵液中pH值与总酸含量预测模型的准确度和稳定性。     

可见/近红外漫反射光谱无损检测甜柿果实硬度

该研究的目的是建立可见/近红外漫反射光谱无损检测甜柿果实硬度的数学模型,评价可见/近红外漫反射光谱无损检测甜柿果实硬度的应用价值。果实硬度采用果皮脆性、果皮强度和果肉平均硬度作为评价指标。在可见/近红外光谱区域(400～2 500 nm),采用改进偏最小二乘法,对比分析了不同导数处理、不同散射及标准化处理的甜柿果实硬度定标模型。结果表明,对于果皮强度和果皮脆性,采用最小偏二乘法、一阶导数处理和标准多元离散校正处理建立的定标模型预测效果较好,RP2分别为0.858和0.862,SEP分别为0.094和0.157,RPD分别为2.47和2.63。对于果肉平均硬度,采用改进偏最小二乘法、一阶导数处理和标准正常化和去散射处理建立的定标模型预测效果较好,RP2为0.82,SEP为0.063,RPD为2.35。因此,可见/近红外漫反射光谱无损检测技术可用于甜柿果实硬度的无损检测。     

The use of elastography to measure quality characteristics of pork semimembranosus muscle

The objectives of this study were to determine if ultrasonic strain image analysis could estimate pork eating quality parameters (such as fresh color, drip loss, and Warner/Bratzler shear). Intact semimembranosus (SM) muscles (cap off) were analyzed for ultimate pH (pH(ult)). Forty-five SM muscles were selected from the larger allotment of fresh hams over a 3-week period. The SM muscles were selected based on high and low pH(ult) in an attempt to represent a wide range of pork quality. Ultrasonic strain images were obtained perpendicular to the SM muscle fibers of an 8-cm cube. Radio-frequency data from each SM were obtained from a field-of-view (FOV) of 40×30 mm(2) and digitized for each compression step. Tissue displacements were computed for each compression step. Tissue strains were computed from displacement data located in the FOV representing areas of harder and softer muscle tissue and converted to gray scale images at 256 levels. Tissue irregularity of hardness and softness was measured using Fractal dimension and Haralicks parameters. Twenty-one Fractal dimension (FR) parameters, at two neighborhood distances (N), from each strain image and nine Haralick's (HAR) textural parameters (inter-pixel distance=1) were analyzed for each image. The variable FR4N4 had a -0.279 correlation with SM ultimate pH (p<0.10); FR6N8 correlated to WB shear force at 0.325 (p<0.05); and FR21N8 had a correlation coefficient of 0.364 with intramuscular fat (p<0.01). Linear regression equations generated from FRN and HAR parameters for intramuscular fat (R(2)=0.468), Warner/Bratzler shear (R(2)=0.360), and 30 h drip loss (R(2)=0.208). Although elastographic measurement was significantly correlated to shear (p<0.05), a better understanding of physical meat texture is necessary before elastography can be used to identify superior quality pork.     

Comparison of Bayesian and partial least squares regression methods for mid-infrared prediction of cheese-making properties in Montbéliarde cows

Assessing the cheese-making properties (CMP) of milks with a rapid and cost-effective method is of particular interest for the Protected Designation of Origin cheese sector. The aims of this study were to evaluate the potential of mid-infrared (MIR) spectra to estimate coagulation and acidification properties, as well as curd yield (CY) traits of Montbéliarde cow milk. Samples from 250 cows were collected in 216 commercial herds in Franche-Comté with the objectives to maximize the genetic diversity as well as the variation in milk composition. All coagulation and CY traits showed high variability (10 to 43%). Reference analyses performed for soft (SC) and pressed cooked (PCC) cheese technology were matched with MIR spectra. Prediction models were built on 446 informative wavelengths not tainted by the water absorbance, using different approaches such as partial least squares (PLS), uninformative variable elimination PLS, random forest PLS, Bayes A, Bayes B, Bayes C, and Bayes RR. We assessed equation performances for a set of 20 CMP traits (coagulation: 5 for SC and 4 for PCC; acidification: 5 for SC and 3 for PCC; laboratory CY: 3) by comparing prediction accuracies based on cross-validation. Overall, variable selection before PLS did not significantly improve the performances of the PLS regression, the prediction differences between Bayesian methods were negligible, and PLS models always outperformed Bayesian models. This was likely a result of the prior use of informative wavelengths of the MIR spectra. The best accuracies were obtained for curd yields expressed in dry matter (CY_DM) or fresh (CY_FRESH) and for coagulation traits (curd firmness for PCC and SC) using the PLS regression. Prediction models of other CMP traits were moderately to poorly accurate. Whatever the prediction methodology, the best results were always obtained for CY traits, probably because these traits are closely related to milk composition. The CY_DM predictions showed coefficient of determination (R²) values up to 0.92 and 0.87, and RS_y,x values of 3 and 4% for PLS and Bayes regressions, respectively. Finally, we divided the data set into calibration (2/3) and validation (1/3) sets and developed prediction models in external validation using PLS regression only. In conclusion, we confirmed, in the validation set, an excellent prediction for CY_DM [R² = 0.91, ratio of performance to deviation (RPD) = 3.39] and a very good prediction for CY_FRESH (R² = 0.84, RPD = 2.49), adequate for analytical purposes. We also obtained good results for both PCC and SC curd firmness traits (R² ≥ 0.70, RPD ≥1.8), which enable quantitative prediction.     

On-line versus off-line NIRS analysis of intact olives

Visible/near-infrared calibrations were developed for the determination of the quality parameters (fat content, moisture and free acidity) of intact olive fruits. The reflectance spectra were acquired in two different instruments (diode-array versus grating monochromator based instruments). The grating monochromator based instrument was used at the laboratory (off-line analysis), whereas the portable diode-array based device was placed on top of a conveyor belt set to simulate measurements in an olive oil mill plant (on-line analysis). Partial least squares (PLS) regression and least squares support vector machine (LS-SVM) were used for the development of the calibration models. A total of 174 samples were prepared for the calibration (N = 122) and validation (N = 52) sets. The root mean square error of prediction (RMSEP) and the residual predictive deviation (RPD) values were better using the diode-array instrument and applying the PLS regression method for the fat content parameter while for the free acidity and moisture content, the LS-SVM algorithm gave the best results. The results obtained seems to suggest the viability of the on-line system, instead of the off-line analysis, for the determination of physicochemical composition in intact olives.     

基于近红外光谱技术快速检测稻谷水分含量

目的:建立一种无损、快速高效的稻谷水分含量检测方法。方法:研究收集了不同年份的稻谷样品161份,运用近红外光谱结合化学计量学方法,通过剔除异常光谱和光谱预处理,采用偏最小二乘法建立稻谷水分含量预测模型。结果:采用主成分分析结合马氏距离的方法剔除异常光谱样品15个,最佳的光谱预处理方式为消除常数偏移量。训练集建立的预测模型(R_CAL²)为0.9943,模型标准偏差(RMSEC)为0.21%,模型交叉验证决定系数(R_CV²)为0.9936,模型交叉验证标准偏差(RMSECV)为0.32%,表明预测模型交叉验证预测样品水分含量准确度高。用验证集样品检验预测模型,模型验证集验证决定系数R 2 VA L为0.9801,模型验证集验证标准偏差(RMSEP)值为0.36%,相对分析误差(RPD)值为7.14,表明预测模型对未知样品的预测准确度高。验证集样品实测值与预测值均值方程T检验结果P值(双侧)为0.879,验证集样品实测值与预测值之间差异不显著,表明预测模型的预测结果可信度高,验证集样品预测值与实测值的误差在±1%,且90%以上的验证集样品其预测值与实测值的误差都在±0.5%以内。结论:建立的稻谷水分预测模型可以实现收储稻谷的无损、快速、准确检测。     

磨盘柿褐变指标的可见/近红外漫反射无损预测研究

为了建立可见/近红外漫反射光谱与磨盘柿果皮和果肉褐变之间的关系,作者在全光谱区域(570～1 848 nm)对比分析了不同处理方法对磨盘柿果皮颜色b*和果肉浊度定标模型的影响。结果表明,应用MPLS、原始光谱和无散射处理建立果皮颜色b*的定标模型预测性能较好,Rp2为0.968,RMSEP为1.417 7,RPD为7.92。应用PLS、一阶导处理和无散射处理建立磨盘柿果肉浊度的定标模型预测性能较好,Rp2为0.757,RMSEP为0.107 9,RPD为2.22。因此,可见/近红外漫反射技术对磨盘柿果皮颜色b*和果肉浊度的快速无损检测具有可行性。     

Development and evaluation of models to predict the feed intake of dairy cows in early lactation

Inaccurate prediction of dry matter intake (DMI) limits the ability of current models to anticipate the technical and economic consequences of adopting different strategies for production management on individual dairy farms. The objective of the present study was to develop an accurate, robust, and broadly applicable prediction model and to compare it with the current NRC model for dairy cows in early lactation. Among various functions, an exponential model was selected for its best fit to DMI data of dairy cows in early lactation. Daily DMI data (n = 8,547) for 3 groups of Holstein cows (at Illinois, New Hampshire, and Pennsylvania) were used in this study. Cows at Illinois and New Hampshire were fed totally mixed diets for the first 70 d of lactation. At Pennsylvania, data were for the first 63 d postpartum. Data from Illinois cows were used as the developmental dataset, and the other 2 datasets were used for model evaluation and validation. Data for BW, milk yield, and milk composition were only available for Illinois and New Hampshire cows; therefore, only these 2 datasets were used for model comparisons. The exponential model, fitted to the individual cow daily DMI data, explained an average of 74% of the total variation in daily DMI for Illinois data, 49% of the variation for New Hampshire data, 67% of the variation for Pennsylvania data, and 64% of the variation overall. Based on all model selection criteria used in this study, the exponential model for prediction of weekly DMI of individual cows was superior to the current NRC equation. The exponential model explained 85% of the variation in weekly mean DMI compared with 42% for the NRC equation. Compared with the relative prediction error of 6% for the exponential model, that associated with prediction using the NRC equation was 14%. The overall mean square prediction error value for individual cows was 5-fold higher for the NRC equation than for the exponential model (10.4 vs. 2.0 kg²/d²). The consistently accurate and robust prediction of DMI by the exponential model for all data-sets suggested that it could safely be used for predicting DMI in many circumstances.