Abstract:

To accurately predict lean meat percentage of commercial pig carcasses in China and to achieve grading on line and fast accounting so as to optimize price and quality, CSB image meater was employed to analyze lean meat percentages, hot carcass weights, back-fat thicknesses and muscle thicknesses of 436 different types of commercial pig carcasses. A regression equation for predicting lean meat percentage of commercial pig carcasses was established based on the data from CSB image meater through multiple linear regression as follows: y = 61.264－0.583x1 + 0.173x2, where x1 was the thickness of the thinnest back fat, and x2 was the vertical distance between the end of gluteus medius muscle and the edge of spinal column, with a determination coefficient R2 of 0.87 and a standard residual error of 2.31%, indicating good degree of fitness. No significant difference between the actual and model-predicted values was observed. Therefore, the established equation has a high accuracy and is suitable to be applied in practice.

Key words: image meater, pig carcass, lean meat percentage, prediction