American Society of Agricultural and Biological Engineers



Click on “Download PDF” for the PDF version or on the title for the HTML version.


If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Application of VIS-SW-NIR Spectroscopy for Detecting Adulteration in Fresh Minced Beef

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org

Citation:  Paper number  131591010 ,  2013 Kansas City, Missouri, July 21 - July 24, 2013. (doi: http://dx.doi.org/10.13031/aim.20131591010 ) @2013
Authors:   Xiaoyu Tian, Yang Xu, Yankun Peng, Xiuying Tang, Shibang Ma
Keywords:   VIS-SW-NIR Spectroscopy Beef Adulteration Prediction Classification

Abstract. The purpose of this study is to evaluate the potential of visible and short wave near infrared (VIS-SW-NIR) spectroscopy, for the rapid, nondestructive detection of beef adulteration. Visible and short wave near infrared (VIS/SW-NIR) spectroscopy was acquired from pure minced beef pork and beef liver, beef and pig fat trimming as well as the mixture samples in the form of minced beef adding different proportion of others, respectively. Partial least squares regression (PLSR) models were established to predict concentration of different kinds adulterants, the correlation coefficient R reached 0.9729, 0.9919, 0.9598, and 0.9445, the root mean square prediction error (RMSEP) were 5.9367, 4.4131, 10.0220 and 10.9009, respectively. According to the average spectral curves of samples with different adulterating level, linear discriminant analysis (LDA) model and support vector machine (SVM) models were built to identify whether the foods were adulterated, and furthermore, classify the level of adulterated level, by selecting some important characteristic wavelength most related to the difference of each adulterants in the minced beef, respectively. The classification accuracy from both types of models was quite high, especially the total prediction accuracy of LDA model reached 94.4, 84.7, 93.3 and 72.2 percent, respectively. The current study demonstrated that the VIS-SW-NIR spectroscopy can be used to detect and classify the amount and level of adulterants added to the minced beef with acceptable precision and accuracy.

(Download PDF)    (Export to EndNotes)