COLOR EVALUATION OF LEAN TISSUE AND FAT OF THE BEEF1

摘要

This study was conducted to develop color evaluation of beef using machine vision system. Machine vision has an advantage to measure larger area than colorimeter and also can be measured another quality factors like distribution of fats. But machine vision measurement is affected by its system component. To measure the beef color with machine vision system, the effect of color balancing control was tested and calibration model of vision system was developed. Neural network for color calibration which learned reference color patches shows the high correlation with colorimeter in L*a*b* coordinates and has an adaptability at various measurement environment. But trained network shows very low correlation with colorimeter when measuring beef color.