Abstract:

This paper took identifying pork and beef by near infrared spectroscopy combined with pattern recognition techniques for purpose. A discriminant analysis model was established based on spread reflection Near-infrared spectroscopy. According to Multiplicative signal correction （MSC）、First derivative and Norris derivative filt to a smooth point ( (5,3) dealed with spectrum . Selecting spectra ranges of 5996.42～5805.84,9058.85～8546.36 cm-1 and adopting TQ Analyst software with Mahalanobis Distance was 1 to set up discriminatory analysis model . The result showed that the model can identify accurately between pork and beef . For this reason , the Near i infrared spectroscopy technology can be used to identify pork and beef .