| China is the largest meat production and consumption country in the world, accounting for aboutone fourth of the consumption and production around the whole world. As a large developing country,China has experienced the30years of reforming and opening-up, and Chinese people started to live withluxuries and have changed the consumption patterns. Meanwhile the requirements of meat in China havealso been changed from quantity to quality. The main factor is quality of meat, determining consumers’purchase intent and market price. To promote meat quality and protect consumers’ interests, we have tocontrol the process of production of meat and meat products. The traditional detection methods of meatquality are the use of chemical analysis, instrumental analysis, sensory analysis and some others’destructive detection methods, which not only consume a lot of time and money, but also could notachieve on-line test. NIR, as a kind of non-destructive, sensitive and rapid detection method, hasreceived more and more people’s concern.NIR is an electromagnetic wave between the visible (VIS) and infrared (IR), which is defined bythe American Society of Materials Testing (ASTM) as the wavelength of780-2526nm and the wavenumber of12820-3959cm-1for spectral region, and developed since1970s of the last century. Now NIRis widely used in various fields, because of its simple sample processing and evaluating in manydifferent meat properties.In this study, beef was used as a main research object and applied the domestic raster scan of nearinfrared spectroscopy instrument (wavelength range1000-1799nm) combined with the chemometricmethods to develop the optimal protocol for near-infrared reflectance spectroscopic on-line evaluation ofbeef quality, determinate the quantitative of meat eating quality factors, predict the grades of beeftenderness, and discriminate the meat from general bull and eliminate cow.The major results and conclusions are summarized as follows:(1)The spectra collection parameters on NIR were analyzed. Through the comparison analysis, theoptimized parameters were chosen, including averaging30spectra per observation, measurement oflongissimus from one side per carcass, and obtaining spectra at a standardized bloom time (2min as soonas possible after ribbing). It is better to build the model under the wavelength1000-1300nm;(2)The models for pH and L~*、a~*、b~*were established by partial least squares regression (PLSR) in1000~1799nm and1000-1300nm. The results demonstrated that the PLSR model of pH are better underthe wavelength1000-1300nm, but the PLSR model of L~*、a~*、b~*is limited, and the NIR models topredict the pH of beef aged for2d are better than the models of beef aging7d;(3)Firstly, we compared the grading result between meat emulsion and meat slice based ontenderness, and then the tenderness of beef aged for2d and7d have been compared. Models of the gradeof beef tenderness after aging for2d and7d were established by using the NIR spectra collected frombeef aged for2d. The results indicated that the NIR spectra collected from beef aged for2d weresuccessfully used to predict the tenderness of beef aged for2d under1000-1300nm, but the use of NIR to predict the tenderness of beef aged for7d need a further research;(4)The influence on NIR spectra was analyzed based on the general bull and eliminated cowcoming from Inner Mongol and Xinjiang. The performance of models for discriminating these two kindsmeat developed by discriminant partial least squares analysis (DPLS) with different spectrapreprocessing methods has been compared. The results in this study suggested that NIR can be used todiscriminate the meat from general bull and eliminate cow. |
PDF Full Text Request