Study On Predtiction For Beef Tenderness

Tenderness, color, flavor and juiciness are the four factors in the consumer perception of beef quality, and the tenderness is the most important one. There are two beef methods for tenderness evaluation including sensory evaluation and assessment instruments. But whatever the sensory evaluation or the instrument assessment, the consumer can not assess tenderness from the sensory of beef, the producers can not predict beef tenderness online and grade beef. So, it urgently requires the development of a new method to predict beef tenderness. Study showed that machine vision method is the most promising approach for prediction of beef tenderness. But all the study focus on the surface texture characteristics of beef to build predictive models of beef tenderness, lacking of theoretical basis. This study aimed to find out the essence characteristics caused different beef texture and then establish the theoretical basis for predicting beef tenderness by the texture. By looking for beef tenderness predictor reflect to the texture, establish the equations, and achieve the prediction for the beef tenderness. This study mainly includes three aspects:(1) the study on biologic factors affecting the beef tenderness of Simmental hybrid bulls;(2) the correlation of shear force values and fascia color values;(3) prediction model of beef tenderness and verification. Content and specific research results are as follows:118,36,54,74months Simmental hybrid bulls were selected from cattle farm of Anhui Da Ming Company. Each group choose6bulls, a total of24. Meat color, fat color, fat content, fiber diameter, connective tissue content, Fiber diameter of the longissimus muscle were measured. The1*; a*; b*of muscle color were37.37,35.82,36.41,38.27;19.31,20.03,20.82,19.81;8.75,9.74,9.35,9.04,respectively. The1*; a*; b*values of fat color were73.63,70.23,75.22,75.50;3.51,2.67,4.67,2.57;9.29,14.19,9.17,13.24, respectively. The fat content were1.79%,7.07%,4.45%,5.81%, respectively. The fiber diameter were44.46μm,47.93μm,50.01μm,56.14μm, respectively. The hydroxyproline content were0.0407%,0.0434%,0.0474%,0.0550%, respectively. The shear force values were5.66kg/cm2,6.11kg/cm2,7.22kg/cm2,8.59kg/cm2, respectively. The study found that the muscle color and fat color values did not change significantly though the different ages. The results showed that the predictors of tenderness having law correlation with the shear force value including:meat color, fat color and fat content, while the R value of muscle fiber diameter, connective tissue content with the shear force values were0.833,0.835, respectively. So, the Fiber diameter and connective tissue content were selected as the predictors of tenderness.2The image acquisition hardware system was established including light sources, cameras, lenses, computer. Used the FCM clustering method to separate the fascia area, The R,G,B values of the fascia were extracted, The R;G;B values of fascia for18,36,54,74months were78.81,83.20,91.44,116.05;64.03,65.85,78.52,86.28;76.38,79.26,85.45,107.33, respectively. The results showed that correlation coefficients of the shear force value with the R, G, B values of fascia were0.856,0.595,0.727. The R value of the fascia had the highest correlation, so, the R value of the fascia was selected as the predictors of tenderness. Also found that the correlation coefficients of the hydroxyproline content with the R,G,B values of fascia were0.735,0.565,0.657. There was a certain degree of correlation.3The multiple linear prediction equation by connective tissue content, fiber diameter, fascia R-value to predict the shear force values was established:Y=-3.376+0.113Xi+58.822X2+0.022X3.(Y was the shear force; Xi was the connective tissue content; X2was the fiber diameter; X3was the fascia R-value). The R2of the prediction model was0.892, highly relevant, and the estimated standard error was0.428. The equation was verified though18beef samples of18-72months from different slaughter plant. The average error of the predicted value and the actual measurement value was0.25kg/cm2, the maximum difference was0.52kg/cm2. The results of the prediction model was satisfactory.