scholarly journals Bread features evaluation by NIR analysis

2008 ◽  
Vol 25 (No. 5) ◽  
pp. 243-248 ◽  
Author(s):  
O. Jirsa ◽  
M. Hrušková ◽  
I. Švec
Keyword(s):  
Cross Validation ◽  
Standard Errors ◽  
Loaf Volume ◽  
Rheological Characteristics ◽  
Average Cell ◽  
Independent Validation ◽  
Predicted Values ◽  
Spectrometer Calibration ◽  
Calibration Equations

Bakery characteristics of wheat dough and the final product and their predictability by NIR technique was investigated using 231 variety and commercial wheat samples (crop years 2003–2005). The behaviour of doughs was assessed with Brabender maturograph and OTG (Germany), the final product quality was evaluated by the baking test and image analysis. NIR spectra of flours were acquired on a NIRSystem 6500 spectrometer. Calibration equations for the selected rheological characteristics were computed by WINISI II using mPLS regression. The quality of prediction was evaluated by means of coefficients of correlation between measured and predicted values from cross and independent validation. A statistically significant dependence (with probability higher than 99%) was found with all rheological characteristics. The standard errors of cross-validation were achieved as follows: dough elasticity 16 BU, bread volume (11 min) 58 BU, specific loaf volume 34 cm<sup>3</sup>/100 g, bread cut area 2.6 cm<sup>2, penetration 4.1 mm, average cell area 0.4 mm<sup>2</sup> and cells per cm<sup>2 7.4.

scholarly journals Characteristics of fermented dough predicted by using the NIR technique

2011 ◽  
Vol 23 (No. 5) ◽  
pp. 184-189 ◽  
Author(s):  
O. Jirsa ◽  
M. Hrušková
Keyword(s):  
Correlation Coefficient ◽  
Cross Validation ◽  
Rheological Characteristics ◽  
Pls Regression ◽  
Independent Validation ◽  
Predicted Values ◽  
Accuracy Of Prediction ◽  
Gas Volume ◽  
Calibration Equations

Rheological characteristics of wheat doughs prepared from 179 variety and commercial wheat flour samples (crop years 1999, 2001, 2002, 2003 and 2004) were assessed with SJA fermentograph (Sweden). NIR spectra of the samples were measured on the spectrometer NIRSystem 6500. Calibration equations with cross and independent validations for all rheological characteristics were computed by NIR Software ISI Present WINISI II using mPLS/PLS regression and ANN. The quality of prediction was evaluated by means of coefficients of correlation between the measured and the predicted values from cross and independent validation. A statistically significant dependence (with probability higher than 99%) was determined in all mentioned rheological characteristics in the case of cross-validation. The results suggested the possibility of prediction particularly for the fermentation gas volume parameter, where a high correlation was achieved (r = 0.888). The accuracy of prediction by independent validation was not found on such a satisfactory level. For the sets of ten commercial flours, higher correlation coefficient was calculated for the final dough volume parameter. &nbsp;

scholarly journals Wheat flour dough alveograph characteristics predicted by NIRSystems 6500

2011 ◽  
Vol 21 (No. 1) ◽  
pp. 28-33 ◽  
Author(s):  
M. Hrušková ◽  
P. Šmejda
Keyword(s):  
Wheat Flour ◽  
Cross Validation ◽  
Rheological Characteristics ◽  
Energy Characteristics ◽  
Coefficient Of Correlation ◽  
Flour Dough ◽  
Independent Validation ◽  
Predicted Values ◽  
Calibration Equations

Rheological quality of wheat dough prepared from 130 variety, stream, and commercial wheat flour samples (wheat harvest 1999, 2000 and 2001) was assessed with alveograph. Spectra of all samples were measured on spectrograph NIRSystems 6500. Calibration equations with cross and independent validation for all rheological characteristics were computed by NIR Software ISI Present WINISI II using MPLS and PLS methods. The quality of the prediction was evaluated by the coefficient of correlation between the measured and the predicted values from cross and independent validation. A statistically significant dependence between the predicted and the measured values (with probability P &lt; 0.01) was determined in P (elasticity) and W (energy) characteristics in the case of cross validation only in the commercial flour sets. P (with P &lt; 0.01) and W (with P &lt; 0.05) were successfully predicted by independent validation in the set of all samples. &nbsp;

scholarly journals Wheat flour dough rheological characteristics predicted by NIRSystems 6500

2013 ◽  
Vol 19 (No. 6) ◽  
pp. 213-218 ◽  
Author(s):  
M. Hrušková ◽  
M. Bednářová ◽  
F. Novotný
Keyword(s):  
Wheat Flour ◽  
Cross Validation ◽  
Rheological Characteristics ◽  
Absorption Time ◽  
Flour Dough ◽  
Independent Validation ◽  
Wheat Flour Dough ◽  
Predicted Values ◽  
Calibration Equations

Rheological quality of wheat dough prepared from 114 wheat flour samples (wheat harvest 1998 and 1999) was assessed by help of farinograph and extensigraph. Spectra of all samples were measured on spectrograph NIRSystems 6500 NIR. Calibration equations with cross and independent validation for all rheological characteristics were computed by NIR Software ISI Present WINISI II using mPLS and PLS methods. The quality of prediction was evaluated by coefficients of correlation between measured and predicted values from cross and independent validation. A statistically significant dependence between predicted and measured values (with probability higher than 99%) was determined in all mentioned rheological characteristics in the case of cross validation. Only farinograph absorption, time of dough development and mixing tolerance were successfully predicted by independent validation. Predictions of extensigraph characteristics were not found out statistically significant probably due to a small number of tested samples.

scholarly journals Prediction of wheat and flour Zeleny sedimentation value using NIR technique.

2011 ◽  
Vol 21 (No. 3) ◽  
pp. 91-96 ◽  
Author(s):  
M. Hrušková ◽  
O. Faměra
Keyword(s):  
Protein Content ◽  
Quality Parameters ◽  
Analytical Quality ◽  
Independent Validation ◽  
Predicted Values ◽  
Accuracy Of Prediction ◽  
Calibration Equations ◽  
Pls Method ◽  
Sedimentation Value

Analytical quality parameters of wheat flour prepared from variety and commercial wheat samples (wheat harvest 1998, 1999, 2000 and 2001) were assessed by means of filter spectrograph Inframatic 8620 ASH (moisture and protein content) and Sedi-tester (Zeleny sedimentation value). The spectra of all samples were measured on spectrograph NIRSystem 6500. Calibration equations with cross and independent validation for all analytical characteristics were computed by NIR Software ISI Present WINISI II using MPLS and PLS method. The quality of prediction was evaluated by SEP and r parameters between the measured and the predicted values from cross and independent validation. In case of Inframatic 8620 ASH, validation was realised by NIRPRG software. A statistically significant dependence between the predicted and the measured values of protein content and Zeleny sedimentation (with probability P &lt; 0.01) was determined in both variety and commercial flour sets in the case of cross and independent validation. Better accuracy of prediction was found with NIRSystem 6500. Both important parameters of wheat were successfully predicted by independent validation with nearly the same accuracy. &nbsp;

scholarly journals Prediction of wheat milling characteristics by near-infrared reflectance spectroscopy

2011 ◽  
Vol 23 (No. 4) ◽  
pp. 145-151 ◽  
Author(s):  
J. Blažek ◽  
O. Jirsa ◽  
M. Hrušková
Keyword(s):  
Near Infrared ◽  
Prediction Models ◽  
Nir Spectroscopy ◽  
Milling Process ◽  
Near Infrared Reflectance ◽  
Independent Validation ◽  
Milling Characteristics ◽  
Predicted Values ◽  
Calibration Equations

The aim of this study was to explore the use of NIR spectroscopy of laboratory milled flour to predict the milling characteristics of wheat. Quantitative traits of the milling process of wheat were predicted by analyses of NIR spectra of six sets consisting of 94 samples. Reference data were obtained by grinding the samples on the laboratory mill Chopin CD1-auto (France), spectral data were measured on spectrograph NIRSystem 6500. Commercial spectral analysis software WINISI II was used to collect spectra, develop calibration equations and evaluate calibration performance. The quality of prediction was evaluated by coefficients of correlation between the measured and the predicted values from cross and independent validation. MPLS/PLS regression and ANN methods were used. A statistically significant dependence (at the probability level of 99%) was determined for all traits studied in the case of cross-validation. Satisfactory accuracy of the prediction models by independent validation was achieved only for semolina extraction rate, models for other characteristics did not show acceptable precision. &nbsp;

True Accuracy of near Infrared Spectroscopy and its Dependence on Precision of Reference Data

2002 ◽  
Vol 10 (1) ◽  
pp. 15-25 ◽  
Author(s):  
L.K. Sørensen
Keyword(s):  
Near Infrared ◽  
Cross Validation ◽  
Reference Data ◽  
Nir Spectroscopy ◽  
Limited Resources ◽  
Standard Errors ◽  
Calibration Data ◽  
Pls Regression ◽  
Calibration Equations ◽  
Very High

A more precise estimate of the accuracy of near infrared (NIR) spectroscopy is obtained when the measured standard errors of cross validation ( SECV) and prediction ( SEP) are corrected for imprecision of the reference data. The significance of correction increases with increasing imprecision of reference data. Very high precision of reference data obtained through replicate analyses under reproducibility conditions may not be the optimal goal for the development of calibration equations. In a situation of limited resources, the precision of the reference data should be related to the obtainable accuracy of the spectroscopic system. Investigation of several routine applications based on the partial least-squares (PLS) regression technique showed that increased precision of calibration data only resulted in marginal improvements in true accuracy if the total standard error of reference results from the beginning was less than the estimated true accuracy of the corresponding NIR calibration.

Fourier Transform near Infrared Spectroscopy Applied to Analysis of Chocolate

2000 ◽  
Vol 8 (4) ◽  
pp. 251-257 ◽  
Author(s):  
Jolana Tarkošová ◽  
Jana Čopíková
Keyword(s):  
Fourier Transform ◽  
Reference Values ◽  
Near Infrared ◽  
Cross Validation ◽  
Nir Spectroscopy ◽  
Sufficient Accuracy ◽  
Standard Methods ◽  
Calibration Models ◽  
Predicted Values ◽  
Calibration Equations

Fourier transform near infrared (FT-NIR) spectroscopy was used to establish calibration equations with the aim of determining sucrose, lactose, fat and moisture in chocolate. The possibility of using FT-NIR spectroscopy for evaluating rheological properties (viscosity and yield) of chocolate was also investigated. The concentrations of individual components and the values of viscosity and yield obtained by standard methods were used as reference values. The spectra of 96 chocolate samples were recorded in reflectance mode in the range of 910–2500 nm using an FT-NIR Nicolet Avatar 360N spectrometer equipped with the UpDRIFT accessory. The first or second derivative transformation of the original NIR spectra gave the best accuracy. A partial least squares (PLS) algorithm was used to create calibration models relating reference values to spectral data. The models were validated using cross-validation. The validation results proved that fat, sucrose and lactose can be predicted with sufficient accuracy, while predicted values for moisture, viscosity and yield are less reliable.

Towards Predicting Relative Belt Edge Endurance With the Finite Element Method

1990 ◽  
Vol 18 (4) ◽  
pp. 216-235 ◽  
Author(s):  
J. De Eskinazi ◽  
K. Ishihara ◽  
H. Volk ◽  
T. C. Warholic
Keyword(s):  
Finite Element Method ◽  
Finite Element ◽  
Three Dimensional ◽  
The Finite Element Method ◽  
Shear Deformations ◽  
Element Analysis ◽  
Vertical Loading ◽  
Predicted Values ◽  
Element Method

Abstract The paper describes the intention of the authors to determine whether it is possible to predict relative belt edge endurance for radial passenger car tires using the finite element method. Three groups of tires with different belt edge configurations were tested on a fleet test in an attempt to validate predictions from the finite element results. A two-dimensional, axisymmetric finite element analysis was first used to determine if the results from such an analysis, with emphasis on the shear deformations between the belts, could be used to predict a relative ranking for belt edge endurance. It is shown that such an analysis can lead to erroneous conclusions. A three-dimensional analysis in which tires are modeled under free rotation and static vertical loading was performed next. This approach resulted in an improvement in the quality of the correlations. The differences in the predicted values of various stress analysis parameters for the three belt edge configurations are studied and their implication on predicting belt edge endurance is discussed.

scholarly journals Development of Machine Learning Models to Predict Compressed Sward Height in Walloon Pastures Based on Sentinel-1, Sentinel-2 and Meteorological Data Using Multiple Data Transformations

2021 ◽  
Vol 13 (3) ◽  
pp. 408
Author(s):  
Charles Nickmilder ◽  
Anthony Tedde ◽  
Isabelle Dufrasne ◽  
Françoise Lessire ◽  
Bernard Tychon ◽  
...  
Keyword(s):  
Machine Learning ◽  
Decision Support ◽  
Support System ◽  
Cross Validation ◽  
Learning Models ◽  
Data Transformations ◽  
Independent Validation ◽  
Sward Height ◽  
Machine Learning Models ◽  
Sentinel 2

Accurate information about the available standing biomass on pastures is critical for the adequate management of grazing and its promotion to farmers. In this paper, machine learning models are developed to predict available biomass expressed as compressed sward height (CSH) from readily accessible meteorological, optical (Sentinel-2) and radar satellite data (Sentinel-1). This study assumed that combining heterogeneous data sources, data transformations and machine learning methods would improve the robustness and the accuracy of the developed models. A total of 72,795 records of CSH with a spatial positioning, collected in 2018 and 2019, were used and aggregated according to a pixel-like pattern. The resulting dataset was split into a training one with 11,625 pixellated records and an independent validation one with 4952 pixellated records. The models were trained with a 19-fold cross-validation. A wide range of performances was observed (with mean root mean square error (RMSE) of cross-validation ranging from 22.84 mm of CSH to infinite-like values), and the four best-performing models were a cubist, a glmnet, a neural network and a random forest. These models had an RMSE of independent validation lower than 20 mm of CSH at the pixel-level. To simulate the behavior of the model in a decision support system, performances at the paddock level were also studied. These were computed according to two scenarios: either the predictions were made at a sub-parcel level and then aggregated, or the data were aggregated at the parcel level and the predictions were made for these aggregated data. The results obtained in this study were more accurate than those found in the literature concerning pasture budgeting and grassland biomass evaluation. The training of the 124 models resulting from the described framework was part of the realization of a decision support system to help farmers in their daily decision making.

Sign in / Sign up

Export Citation Format

Share Document