Sage Journals: Discover world-class research

Abstract

In this paper, we present an approach for incorporating chemical band assignment information in regression models between spectra and constituents. It is shown how the matrices in this L-shaped data structure can be combined and give direct information of the relationships between theoretical chemical band assignment, spectral wavelengths and the responses. The chosen application is NIR spectroscopic measurements of canola seeds. Variable selection based on partial least squares regression using jack-knifing within a cross-model validation (CMV) framework is applied for removing non-relevant spectral regions. Extended multiplicative scatter correction was applied as a spectral pre-treatment to remove physical scatter effects in the spectra. The results show a high degree of correspondence between the objectively found wavelength bands from CMV and the reported chemical interpretation found in the literature.

Keywords

PLS regression cross-model validation band assignment extended multiplicative scatter correction fatty acids interpretive spectroscopy

Get full access to this article

View all access options for this article.

References

Hjort

, Computer Intensive Statistical Methods. Chapman and Hall, London, UK, p. 40 (1994).

Anddrssen

Dyrstad

Westad

and Martens

, “Reducing over-optimism in variable selection by cross-model validation”, Chemometr. Intell. Lab. Syst. 84(1/2), Special issue. p. 69 (2006).

Nørgaardd

and Bro

, Proceedings of International PLS 99, CISIA – CERESTA, France, p. 187 (1999).

Westad

Afanth

N.K.

and Bro

, “Finding relevant spectral regions between spectroscopic techniques by use of cross model validation and partial least squares regression”, Anal. Chim. Acta 595, 323 (2007). doi: 10.1016/j.aca.2007.02.015

Westad

and Martens

, “Variable selection in near infrared spectroscopy based on significance testing in partial least squares regression”, J. Near Infrared Spectrosc. 8, 117 (2000).

Martens

Anderssen

Flatberg

Gidskehaug

L.H.

Høy

Westad

Thybo

and Martens

, “Regression of a data matrix on descriptors of both its rows and of its columns via latent variables: L-PLSR”, Comput. Stat. Data Anal. 48(1), 103 (2005). doi: 10.1016/j.csda.2003.10.004

Martens

Nielsen

Pram J.

and Ellingsen

Balling S.

, “Light scattering and light absorbance separated by extended multiplicative signal correction. Application to near-infrared transmission analysis of powder mixtures”, Anal. Chem. 75(3), 394 (2003). doi: 10.1021/ac020194w

Workman

, “Interpretive spectroscopy for near infrared”, Near infrared spectroscopy: The future waves, Ed by Davies

A.M.C.

and Williams

Phil

, NIR Publications, Chichester, West Sussex, UK, p. 6 (1996).

Siesler

, “Quality control and process monitoring by vibrational spectroscopy”, NIR news 11(3), 9 (2000).

10.

Hourant

Baetten

Morales

M.T.

Meurens

Aparicio

, “Oil and fat classification by selected bands of NIRS”, Appl. Spectrosc. 54(8), 1168 (2000). doi: 10.1366/0003702001950733

11.

Snieder

Dekker

N.A.G.

Wiedemann

S.C.C.

and Hansen

W.G.

, “Determination of very low unsaturation levels in oleochemical by NIR”, in Near infrared spectroscopy: Stretching the NIR-spectrum to the limit, Ed by Davies

A.M.C.

and Garrido-Varo

, NIR Publications, Chichester, West Sussex, UK, p. 999 (2004).

12.

AOCS Standard Procedure for Iodine Value (IV), ABB Application Note (2004).

13.

Behmer

Montasell

and Pascual

Villar C.

, “Applications with a Mediterranean flavour: Quality control of olives and olive oil with FTNIR”, in Near infrared spectroscopy: NIR in action—making a difference, Ed by Burling-Claridge

G.R.

Holroyd

S.E.

and Sumner

R.M.W.

New Zealand NIRS Society Inc., Hamilton, New Zealand, p. 330 (2007).

14.

Guide for Infrared Spectroscopy, p.4 http://www.bruke-roptics.de/downloads/index.html.

Incorporating Chemical Band-Assignment in near Infrared Spectroscopy Regression Models

Abstract

Keywords

Get full access to this article

References