Estimation of carcass composition by computer image analysis in the cross sections of cross-bred steers

Estimation equations for carcass composition were obtained using the information extracted from the carcass cross section by Computer Image Analysis (CIA). The total kilograms of lean, fat, and bone, and their percentages, were measured on the left side of the carcasses of F1 (cross-bred between Japanese Black and Holstein) steers by physical dissection. Traced data of the cross section between the 5th and 6th ribs (Data set I) and pictures of carcass cross section between the 7th and 8th ribs (Data set II) were subjected to image analysis. Various information on both the individual muscles and the overall outline of the cross section was extracted by the CIA technique. Maximum R2 improvement method of the stepwise procedure was used to choose the best regression equation to estimate carcass composition as total kilograms and percentages of lean, fat, and bone. The data sets were also adjusted for age and the stepwise procedure was also conducted. Coefficients of determination, adjusted for the degrees of freedom (adjusted R2) of the regression equations for estimating carcass composition, were high, i.e., 0.779 to 0.959 for kilograms of lean, fat, and bone, whereas for the percentages of lean, fat, and bone were high, i.e., 0.788 to 0.952, respectively. For the adjusted data, the adjusted R2 for estimating kilograms of lean, fat, and bone with Data sets I and II were 0.729, 0.633, and 0.598, and 0.813, 0.806, and 0.878, respectively, while for the percentages of lean, fat, and bone were 0.793, 0.623, and 0.378, and 0.953, 0.989, and 0.467, respectively. When the estimation equation obtained from the unadjusted Data set I was fitted with the information extracted from Data set II, the correlation coefficients between the values estimated by the equation and the values obtained by physical dissection on carcass composition were high, ranging from 0.70 to 0.92. On the other hand, the correlation coefficients obtained from the adjusted data sets were low. Key words: Estimation equation, computer image analysis, carcass composition, carcass cross section, F1 steers