Share Email Print
cover

Proceedings Paper

Integration of independent component analysis with near infrared spectroscopy for evaluation of rice freshness
Author(s): Yung-Kun Chuang; Suming Chen; Stephen R. Delwiche; Y. Martin Lo; Chao-Yin Tsai; I-Chang Yang; Yi-Ping Hu
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Determination of freshness is an important issue for rice quality. Near infrared (NIR) spectroscopy, a rapid nondestructive inspection method based on specific absorptions within a given range of wavelengths corresponding to the constituents in the sample, has been widely applied for evaluation of internal quality of agricultural products. Since NIR spectra of a mixture may be approximated as the linear addition of individual spectra of the constituents in the mixture, such a mixture spectrum thus can be regarded as 'blind sources' as the proportion of constituents in the samples remains unknown. A multiuse statistical approach, independent component analysis (ICA), is capable of disassembling the mixture signals of Gaussian distribution into non-Gaussian independent constituents, and (with assumption of independent constituent spectral response) can give a complete explanation about the property of constituents in the mixture. By example, a total of 180 white rice samples were collected from 6 crop seasons (from 2006 to 2010) for the purpose of developing an ICA NIR based procedure for rice freshness. , Values of pH were determined by a conventional (bromothymol blue methyl red) method. The calibration model of white rice yielded Rc = 0.939, SEC = 0.202, rp = 0.803 and SEP = 0.233 using original full wavelength range (400 to 2498 nm) spectra and 5 independent components (ICs). Freshness of the white rice can be distinguished either visually by 3-dimensional diagram composed from ICs 2, 3 and 4, or statistically by a calibration model. The results show that ICA with NIR can quickly identify and effectively quantify the pH value in white rice with high predictability, and has the potential to be a useful tool for evaluating rice freshness.

Paper Details

Date Published: 5 May 2012
PDF: 7 pages
Proc. SPIE 8369, Sensing for Agriculture and Food Quality and Safety IV, 83690X (5 May 2012); doi: 10.1117/12.921309
Show Author Affiliations
Yung-Kun Chuang, National Taiwan Univ. (Taiwan)
Univ. of Maryland, College Park (United States)
USDA Agricultural Research Service (United States)
Suming Chen, National Taiwan Univ. (Taiwan)
Stephen R. Delwiche, USDA Agricultural Research Service (United States)
Y. Martin Lo, Univ. of Maryland, College Park (United States)
Chao-Yin Tsai, National Taiwan Univ. (Taiwan)
I-Chang Yang, Univ. of Maryland, College Park (United States)
USDA Agricultural Research Service (United States)
Yi-Ping Hu, National Taiwan Univ. (Taiwan)


Published in SPIE Proceedings Vol. 8369:
Sensing for Agriculture and Food Quality and Safety IV
Moon S. Kim; Shu-I Tu; Kuanglin Chao, Editor(s)

© SPIE. Terms of Use
Back to Top