Prediction of Pork Quality Characteristics Using Visible and Nearinfrared Spectroscopy

Visible/nearinfrared (Vis/NIR) spectroscopy was tested as a nondestructive method to assess the quality of freshpork loin. Quality characteristics of pork were measured to develop calibration models for a diodearray Vis/NIR reflectancespectrophotometer. Spectral measurements and measurements for pH, waterholding capacity (WHC), XYZ tristimulusvalues, cook yield, shear force, proximate analysis (moisture, fat, and protein content), and subjective color, marbling, andfirmness scores were performed for 411 fresh pork loin samples. Approximately 80% of the sample set was used for calibrationand 20% for validation. Partial least squares (PLS) calibration models were developed for each characteristic measured. XYZtristimulus values were predicted well with R2 > 0.91 and SECV < 0.96 for calibration, and R2 > 0.88 and SEP < 1.2 forvalidation. Moisture, fat, and protein content were predicted less well with R2 > 0.69 and SECV < 0.007 for calibration, andR2 > 0.61 and SEP < 0.007 for validation. The calibrations for the other characteristics (shear force, cook yield, pH,waterholding capacity, and scores for color, marbling, and firmness) varied from marginally good to poor with R2 valuesranging from 0.584 to 0.002.