Detection of pork freshness using NIR hyperspectral imaging

摘要

In this study the potential of a pushbroom hyperspectral imaging system in the near-infrared (NIR) range (900-1700 nm) as a non-destructive technique was investigated for the detection of freshness in pork meat. Pork samples from the longissimus dorsi muscle of different quality grades were investigated. Each sample was imaged by the hyperspectral imaging system at 2 days post-mortem and then kept frozen for six months. Samples were then removed from the freezer, thawed and imaged again by the system. The acquired hyperspectral images for both fresh and thawed samples were calibrated, the main region of interest (ROI) was identified and the spectral data were extracted. The spectral data were analyzed by multivariate analysis and the most important wavelengths were identified by 2nd derivative of the raw spectra. Results revealed that hyperspectral imaging technique has an excellent ability as a fast and non-destructive method for fresh pork authentication.