Image Analysis Prediction of Beef Carcass Composition from the Carcass Section

用图像分析仪对牛肉屠体的5~6肋骨横截面进行了分析。对预测屠体成分有用的测量方法进行了评价。测量总横截面、肌肉和脂肪面积的面积、环长、长、短轴长度和重心。以重复性和两次测量的变异系数作为方法的精密度指标。通过逐步回归分析，筛选出预测屠体组成的最佳回归方程，即总千克重、瘦肉率、脂肪率和骨质率。描记得到的实际面积与图像分析仪得到的实际面积之间的相关性是+。88至+。98.重复性从+到+。89到+。99.预测瘦肉率和脂肪率的最重要变量是皮下脂肪面积百分比，而预测瘦肉量和脂肪总公斤的最重要变量是各组织面积。在预测骨骼总重时，肌肉重心之间的距离是一个重要的自变量。预测日本黑猪瘦肉率的最重要变量是总面积或脂肪面积(cm^2)，预测脂肪或骨骼百分比的最重要变量是脂肪面积百分比。经回归方程调整自由度(R^2)的决定系数分别为0.727、0.864和-0.905。另一方面，预测总死亡率的最重要变量是总面积(cm^2)。在预测脂肪和骨骼的总公斤时，肌肉中心之间的距离是一个重要的自变量。R^2一度高达0.9左右。

The 5-6th rib cross section of beef carcasses was analyzed by the Image Analyzer. The measurements useful-for the prediction of carcass composition were evaluated. The measurements were area, circular length, long and short axis length and the center of gravity of the total cross section, several muscles and fat area. The repeatability and coefficient of variation of measures repeated twice by the I'mage Analyzer were used as the index of precision of the method. Stepwise regression analysis was used to choose the best regression equation to predict carcass composition as total kilograms and percentages of lean, fat and bone.1. The correlations between actual area done by tracing and those done by the Image Analyzer were from +. 88 to +. 98. The repeatabilities ranged from +. 89 to +. 99. The most important variable to predict the percentage of lean and that of fat was subcutaneous fat area percentage, while to predict total kilograms of-lean and fat was each tissue area. In predicting total kilograms of bone the distance between the centers of gravity of muscles was an important independent variable.2. The most important variable to prediet the percentage of lean in the Japanese Black was total area or fat area(cm^2), while that to predict the percentages of fat or bone was fat area percentage. Coefficents of determination adjusted for the degrees of freedom(R^2)by the regression-equations for the percentages of lean, fat and bone were 0.727, 0.864 and 0.905, -respectively. On the other hand, the most important variable to predict total kil(igrams of lean, fat arid bone was total area(cm^2). In predicting total kilograms of fat and bone, the distance between the centers of muscles was an important independent variable. The R^2 were as high as around 0.9.