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.

Prediction of the total viable bacterial count based on visible/near infrared spectra during pork spoiling

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

Citation:  Paper number  131593677,  2013 Kansas City, Missouri, July 21 - July 24, 2013. (doi: @2013
Authors:   Xiuying Tang, He Guo, Yankun Peng, Yang Xu, Jie Zheng
Keywords:   Pork Spoilage Total viable bacterial count (TVC) Visible/near infrared spectra Partial least squares regression (PLSR) .

Abstract. Total viable bacterial count (TVC) is one of the most important parameters during pork spoiling, by which we can have a knowledge of the degree of spoilage of pork. In this study, we applied visible/near infrared spectroscopy technology to get the reflectance spectra of pork stored under 4℃, and then the TVC had been calculated. The experiment of detection lasted for 13 days which contained the whole process of the pork spoiling. The wavelength range selected was from 460nm to 940nm. The original data of reflectance spectra were processed by standard normal variables transform (SNV) to remove the noise, and eventually the prediction model of TVC of pork was established by using partial least squares regression (PLSR) method. We compared the results of these models established by these different wavelengths chosen by different wavelength choosing methods. The results showed that the prediction model which was established with the wavelengths selected by repeated Multiple Linear Regression (MLR) choosing method was better than others. Its correlation coefficient of calibration (Rc) was 0.9688, standard error of calibration (SEC) was 0.5662, correlation coefficient of validation (Rv) was 0.9711, standard error of validation (SEV) was 0.6173. And in the ten wavelengths chosen by this method, there are three, 547.33 nm, 579.33 nm, 579.53nm, were close to the wavelengths chosen according to the feature points of the reflectance spectrum curve.

(Download PDF)    (Export to EndNotes)