Evaluation of Pork Loin Freshness Using Absorbance Characteristic of Near-Infrared

摘要

Near-infrared spectroscopy (NIRS) is a promising technique for large-scale meat quality evaluation. Freshness, as well as meat quality, is an important parameter for consumers in purchasing this product. In this study, near-infrared (NIR) spectra in the range 950-1650 nm were used for freshness measurement of pork loin. Three parameters, volatile basic nitrogen (VBN), thiobarbituric acid (TBA), and total microbes (TM), were used as indicators to estimate the freshness. Fifty loin muscle samples from ten independent batches were stored at 5�C � 1�C. The freshness parameters and NIR spectra were measured during 18 days of storage at 4-day intervals. Pre-processing of NIR spectra was conducted before statistical analysis. Principal component regression (PCR) and partial least square regression (PLSR) were used as calibration and validation for models to predict the parameters from the near-infrared spectra. Concerning the number of variables used for model development, the PLSR models were reasonable in estimating VBN, TBA, and TM of pork. The coefficients of determination between spectra and each parameter using PLSR were 0.8212, 0.5895, and 0.7804 with standard errors of cross-validation (SECV) of 2.11, 0.13, and 0.79, respectively, for VBN, TBA, and TM.