Premium
Quantifying Surface Lipid Content of Milled Rice via Visible/Near‐Infrared Spectroscopy
Author(s) -
Chen H.,
Marks B. P.,
Siebenmorgen T. J.
Publication year - 1997
Publication title -
cereal chemistry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.558
H-Index - 100
eISSN - 1943-3638
pISSN - 0009-0352
DOI - 10.1094/cchem.1997.74.6.826
Subject(s) - calibration , partial least squares regression , chemistry , analytical chemistry (journal) , standard error , infrared , infrared spectroscopy , near infrared spectroscopy , second derivative , wavelength , spectroscopy , biological system , optics , mathematics , chromatography , statistics , physics , mathematical analysis , organic chemistry , quantum mechanics , biology
Visible/near‐infrared calibrations were developed and tested for surface lipid content (SLC) of milled long‐grain rice. Three rice varieties were divided into two sample sets, with one containing two variables (degree of milling and variety) and another containing three variables (degree of milling, variety, and kernel thickness). The reflectance calibration equation from the set with three variables was much more accurate in predicting SLC than was the calibration from the two‐variable set. Optimal calibration and prediction were obtained by combining both visible and near‐infrared wavelength ranges and using the modified partial least squares technique on spectra pretreated by standard normal variate and first derivative methods. The best calibration yielded a coefficient of determination ( R 2 ) of 0.99 and a standard error of prediction of 0.04% SLC, which was approximately 1.5 times the standard error of calibration and also 1.5 times the SLC measurement error.