Prediction of drip-loss, pH, and color for pork using a hyperspectral imaging technique

Many subjective grading methods with poor repeatability and tedious procedures are still widely used in meat industry. In this study, a hyperspectral-imaging-based technique was investigated to evaluate its potentials for objective determination of pork quality attributes. The system extracted spectral and spatial characteristics simultaneously to determinate the quality attributes, drip loss, pH, and color, of pork meat. Six feature band images were selected for predicting the drip loss (459, 618, 655, 685, 755 and 953nm), pH (494, 571,637, 669, 703 and 978nm) and color (434, 494, 561, 637, 669 and 703nm), respectively. Two intensity indices of the band images were used as inputs to establish neural network models to predict the quality attributes. The results showed that with the hyperspectral-imaging system, the drip loss, pH, and color of pork meat could be predicted with correlation coefficients of 0.77, 0.55 and 0.86, respectively. Pork meat could be classified based on their exudative characteristics and color successfully.