Application of Hyperspectral Imaging Technique for Determination of Pork Quality Attributes

Meat grading has always been a research subject because of its economical importance and the large variations among meat product qualities. In this study, a hyperspectral imaging system in the near-infrared (NIR) range (900-1700 nm) was developed for quality assessment of pork meat. Pork samples were classified in three quality grades, as reddish-pink, firm and non-exudative (RFN), pale, soft and exudative (PSE), and dark, firm and dry (DFD) based on colour, texture and exudation of the meat. Spectral information obtained has shown that there are differences among pork meat quality classes that allows for classification of samples based on spectral features. Some significant wavelengths linked to drip loss, pH and colour attributes were identified by using first derivative of the spectra. Principal component analysis (PCA) has shown that spectral information can distinguish pork meat according to quality characteristics. Results are encouraging and show the promising potential of hyperspectral technique applications for fast and non-destructive assessment of pork quality.