Quantifying Total Viable Count in Pork Meat Using Combined Hyperspectral Imaging and Artificial Olfaction Techniques

Abstract

Total viable count (TVC) of bacteria is one of the most important indexes in evaluation of quality and safety of meat. This study attempts to quantify the TVC content in pork by combining two nondestructive sensing tools of hyperspectral imaging (HSI) and artificial olfaction system based on the colorimetric sensor array. First, data were acquired using HSI system and colorimetric sensors array, respectively. Then, the individual characteristic variables were extracted from each sensor. Next, principal component analysis (PCA) was used to achieve data fusion based on these characteristic variables from two different sensor data for further multivariate analysis. In developing the models, linear (PLS and stepwise MLR) and nonlinear (BPANN and SVMR) pattern recognition methods were comparatively employed, and they were optimized by cross-validation. Compared with other models, the SVMR model achieved the best result, and the optimum results were achieved with the root mean square error of prediction (RMSEP) = 2.9913 and the determination coefficient (R p ) = 0.9055 in the prediction set. The overall results showed that it has the potential in nondestructive detection of TVC content in pork meat by integrating two nondestructive sensing tools of HSI and colorimetric sensors with SVMR pattern recognition tool.