Real-time Evaluation System of Pork Freshness based on Multi-channel Near-infrared Spectroscopy

Pork freshness influences the consumers' health and has become a food safety issue. The quality of pork meat can be determined in terms of its freshness, which influences the purchasing decision of the meat by consumers. The pork quality in general is recognized by its appearance and smell, which is unscientific and subjective method. As such, real-time, non-destructive, rapid and accurate detection of pork freshness is current industrial desire. A multi-channel near-infrared spectroscopy in the wave length range of 300 to 2500nm was used to detect three parameters of pork freshness namely, Total Volatile Basic-Nitrogen (TVB-N), pH value and meat color. The system mainly included hardware system and software system. The hardware system mainly included the visible and near infrared spectroscopy system, signals transmission system and transfer system. Serial signal detection, spectra collecting, data saving and processing, model predicting was planted into the software system. A controlled mechanical system was fabricated to hold the detection device and acquire the NIR spectral data. A Partial Least Square Regression (PLSR) method was used to develop prediction model of pork freshness. The research shows that the multi-channel near-infrared can be a mile-stone for development of real-time, non-destructive, rapid and accurate detection technology for accessing pork quality.