Theory and application of near infrared reflectance spectroscopy in determination of food quality

Cen, Haiyan; He, Yong

doi:10.1016/j.tifs.2006.09.003

Cited by 798 publications

(422 citation statements)

References 96 publications

Supporting

Mentioning

397

Contrasting

Unclassified

Order By: Relevance

“…The effect (i.e., influence) of each preprocessing step can be calculated by taking the average of the response variable of this particular step at the high level (y ̅ + ) minus the average response variable at the low level (y ̅ − ): = ̅ − ̅ + − y y effect (2) A negative effect indicates that y ̅ + is smaller than y ̅ − , i.e., performing a specific step has led to a decrease in the response variable and thus improved model performance. The complete design matrix is shown in the first five columns of Table 2.…”

mentioning

confidence: 99%

Simple and Effective Way for Data Preprocessing Selection Based on Design of Experiments

et al. 2015

View full text Add to dashboard Cite

The selection of optimal preprocessing is among the main bottlenecks in chemometric data analysis. Preprocessing currently is a burden, since a multitude of different preprocessing methods is available for, e.g., baseline correction, smoothing, and alignment, but it is not clear beforehand which method(s) should be used for which data set. The process of preprocessing selection is often limited to trial-and-error and is therefore considered somewhat subjective. In this paper, we present a novel, simple, and effective approach for preprocessing selection. The defining feature of this approach is a design of experiments. On the basis of the design, model performance of a few well-chosen preprocessing methods, and combinations thereof (called strategies) is evaluated. Interpretation of the main effects and interactions subsequently enables the selection of an optimal preprocessing strategy. The presented approach is applied to eight different spectroscopic data sets, covering both calibration and classification challenges. We show that the approach is able to select a preprocessing strategy which improves model performance by at least 50% compared to the raw data; in most cases, it leads to a strategy very close to the true optimum. Our approach makes preprocessing selection fast, insightful, and objective.

show abstract

mentioning

confidence: 99%

Simple and Effective Way for Data Preprocessing Selection Based on Design of Experiments

et al. 2015

View full text Add to dashboard Cite

show abstract

“…Samples for the calibration series were prepared by adding free oleic acid to the ground seeds characterized with known content of oleates [8]. For this purpose, a solution of oleic acid in a volatile organic solvent was evenly sprayed over the surface of the sample stirring it.…”

Section: Preparation Of the Calibration Series Of Raw Materials Samplesmentioning

confidence: 99%

“…The procedure for analyzing all samples involves recording the spectrum of the standard that is in the instrument, recording the spectrum of the prepared samples that are under the study, and processing the obtained results with the appropriate software [8]. The time for recording of the spectrum of one sample does not exceed 2 minute.…”

Section: Spectra Collection and Pretreatmentmentioning

confidence: 99%

Determination of oleic acid in the samples of sunflower seeds by method of nir-spectroscopy

Hutsalo¹,

Mank²,

Kovaleva³

2017

Ukr. food j.

View full text Add to dashboard Cite

Introduction. The possibility of using NIR spectroscopy to determine oleic acid in sunflower seeds has not been studied, so research on this field is perspective.Materials and methods. The spectra of seed samples of various sunflower varieties with a known content of oleic acid and the same samples additionally enriched with oleic acid were investigated by NIR diffusion reflectance spectroscopy with using the instrument "Infrapid-61". To process the results obtained, the methods of mathematical analysis were applied.Results and Discussion. In the NIR spectra of samples of dried sunflower seeds, in comparison with the spectra of raw seeds, the expected decrease in the coefficient of diffuse reflection is observed in the range 1920-1940 nm related to the moisture content in the sample. Analysis of NIR spectra of a calibration series of dried seeds enriched with oleic acid shows an increase in the coefficient of diffuse reflection in the wavelength ranges 1920-1940 nm and 2140-2160 nm in proportion to the growth of the mass portion of oleic acid. Corresponding calculations, calibration curves and the obtained equation describing the dependence demonstrate a linear dependence of the reflection coefficient on the mass portion of oleic acid in the sample at the wavelength of 2140 nm with a confidence level of 98%. The dependency found can be used for the quantitative determination of oleic acid in a sunflower seed sample of unknown composition. By the magnitude of the coefficient of diffuse reflection of the sunflower seed sample containing unknown amount of oleic acid, its mass portion in the sample can be determined from the graph. The diffuse reflectance spectra of husked and crushed seed samples and corresponding spectra of crushed sunflower seeds with husks containing the same oleate amount are practically identical with the wavelength range 1330-2370 nm. Therefore, this method can be used for the analysis both intact seeds and seeds separated from husks. The express method of diffuse reflection of NIR spectroscopy can be considered as an alternative to chemical methods for determining the quality indicators of fatcontaining raw materials.Conclusions. The method of NIR spectroscopy is perspective for the determining of other fatty carboxylic acids in sunflower seeds.

show abstract

“…Examples in literature of engineering problems solved using classification include solving constrained optimization problems (Basudhar et al [2012], Handoko et al [2008]), finding quasi-optimal regions (Singh et al [2013]), determining food quality (Cen and He [2007]), measuring analog circuit performance (De Bernardinis et al [2003]), detecting faults in aircraft engines (Rausch et al [2004]), etc.…”

Section: Classification and Simulation-based Engineeringmentioning

confidence: 99%

A sequential sampling strategy for adaptive classification of computationally expensive data

Singh

Herten

Deschrijver

et al. 2016

Struct Multidisc Optim

View full text Add to dashboard Cite

Many real-world engineering problems can be represented and solved as classification problems. These problems are typically encountered in design optimization and use expensive data, since the data are often sourced from computationally expensive simulations. It is therefore crucial to solve the classification problem using as little data as possible. This necessitates an iterative classifier construction procedure beginning with a very small training set, which is supplemented in each iteration by a small batch of new data points. The sequential sampling algorithm selects locations in the input space and the simulator calculates the corresponding class label. This paper describes a sequential sampling algorithm for adaptive classification of expensive data.

show abstract

Theory and application of near infrared reflectance spectroscopy in determination of food quality

Cited by 798 publications

References 96 publications

Simple and Effective Way for Data Preprocessing Selection Based on Design of Experiments

Simple and Effective Way for Data Preprocessing Selection Based on Design of Experiments

Determination of oleic acid in the samples of sunflower seeds by method of nir-spectroscopy

A sequential sampling strategy for adaptive classification of computationally expensive data

Contact Info

Product

Resources

About