A Method for Wavelength Standardisation in Filter Instruments

Fearn, Tom; Eddison, C G; Withey, Robin P.; Cowe, Ian A.

doi:10.1255/jnirs.82

Cited by 8 publications

(6 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3 Different authors have recently studied the interest of these respective methods for multivariate calibration in chemometrics. [4][5][6] The rationale behind PCR, PLS and RR is to reduce the variability of the ␤ estimations by shrinking the coefficient vector away from the OLS solution. 4,7 PCR and PLS are discrete shrinkage methods because the parameter of interest is the number of latent variables introduced in the model, whereas RR is a continuous shrinkage method as it depends on a scalar parameter k. In PCR the latent variables are chosen among the principal components of X and in PLS the latent variables included in the model are iteratively computed using information from both X and y.…”

Section: Introductionmentioning

confidence: 99%

Principal component regression, ridge regression and ridge principal component regression in spectroscopy calibration

Vigneau

Devaux²,

Qannari

et al. 1997

J. Chemometrics

View full text Add to dashboard Cite

SUMMARYRidge regression (RR) and principal component regression (PCR) are two popular methods intended to overcome the problem of multicollinearity which arises with spectral data. The present study compares the performances of RR and PCR in addition to ordinary least squares (OLS) and partial least squares (PLS) on the basis of two data sets. An alternative procedure that combines both PCR and RR is also introduced and is shown to perform well. Furthermore, the performance of the combination of RR and PCR is stable in so far as sufficient information is taken into account. This result suggests discarding those components that are unquestionably identified as noise, when the ridge constant tackles the degeneracy caused by components with small variances.

show abstract

Section: Introductionmentioning

confidence: 99%

Principal component regression, ridge regression and ridge principal component regression in spectroscopy calibration

Vigneau

Devaux²,

Qannari

et al. 1997

J. Chemometrics

View full text Add to dashboard Cite

show abstract

“…for the particular bias correction between spectral responses of the slave and master instruments, equations (9) and (10) are still valid with m i = 1.…”

Section: Standardising the Spectra Between Master And Slave Instrumentsmentioning

confidence: 99%

“…[7][8][9] essentially, there are two main approaches in calibration transfer procedures: standardisation of the spectral responses or standardisation of the predicted values. 5,10 In the first case, we aim at finding the relationship between the absorbances of a fixed sample recorded on different instruments so that the measured spectra could be transformed Introduction into the spectra recorded in a master instrument, whose calibration equations are then applied. In the second case, we aim at finding, for each property and material, the relationship between the predicted value of the master instrument and the value predicted with another instrument by using the raw master equation.…”

mentioning

confidence: 99%

Working with a Set of Filter near Infrared Instruments

Gatius

Llovera

Lloveras

et al. 2011

Journal of Near Infrared Spectroscopy

View full text Add to dashboard Cite

There is an increasing need for using a set of near infrared (NIR) instruments in industrial activities. In this study, three methodologies of working with a set of NIR instruments are reviewed: the standardisation of the predicted values, the standardisation of the absorbances and the modelling of the instrument differences. Modelling the NIR differences is performed by including in the calibration/validation sets spectra recorded in any of the different instruments of the set without any pre-treatment or after transfer by orthogonal projection (TOP) pre-transformation. In the latter case, TOP calculations from different sets of products are assayed. The methods are briefly presented and discussed through the results of a set of four InfraAlyzer 2000 instruments, 19-filter spectrometers which are used in the quality control of fresh alfalfa. Although the optimal procedure depends on the particular conditions of the industrial work (number of products, number of properties, number of instruments, etc.), all the above methods have been found to be suitable for operating with a set of NIR instruments. The present results also indicate an independence of the TOP results from the product used to compute the TOP factors, at least for the present set of NIR instruments which all belong to the same model and company.

show abstract

“…This is easier to demand than to achieve, however, and the chemometric methods for calibration transfer described here are likely to be needed for some time to come. Adhihetty et al 1 42 describe a wavelength standardisation method for filter instruments that corrects for differences between filters.…”

Section: Some Solutionsmentioning

confidence: 99%

Standardisation and Calibration Transfer for near Infrared Instruments: A Review

Fearn

2001

Journal of Near Infrared Spectroscopy

Self Cite

221

178

View full text Add to dashboard Cite

Transferring calibrations between near infrared instruments is not always straightforward, even when the instruments are nominally the same. Problems can include both wavelength shifts and differences in absorbance response between instruments. This review describes a number of chemometric methods that have been developed to aid calibration transfer. The approaches are classified under three headings: making robust calibrations, adjusting calibrations and adjusting spectra. Calibrations can be made more robust, and more transferable, by the appropriate use of spectral pre-treatments that reduce between-instrument variability. Robust calibrations can also be made by pooling data from several instruments when calibrating. Calibrations can be adjusted using skew and bias corrections estimated from a modest number of samples with known reference values. Under the third heading, adjusting spectra, come methods such as piecewise direct standardisation and the patented method of Shenk and Westerhaus that use spectral information from samples measured on two instruments to match the spectral response of the instruments.

show abstract

A Method for Wavelength Standardisation in Filter Instruments

Cited by 8 publications

References 4 publications

Principal component regression, ridge regression and ridge principal component regression in spectroscopy calibration

Principal component regression, ridge regression and ridge principal component regression in spectroscopy calibration

Working with a Set of Filter near Infrared Instruments

Standardisation and Calibration Transfer for near Infrared Instruments: A Review

Contact Info

Product

Resources

About