Abstract
This paper proposes a new estimation method based on a neural network modeling using dynamic texture features extracted from ultrasound image to estimate Beef Marbling Standard (BMS) number accurately. The correlation coefficient between estimated and actual BMS numbers was r=0.75(P<0.01) by leave-one-out method using dynamic texture features. This result showed that dynamic texture features extracted from dynamic image were effective in the accurate estimation of BMS number.
