Estimation of Beef Marbling Standard for Live Cattle Using Multi-input CNN with Ultrasound Images and Individual Cattle Information

There is great interest among livestock farmers in estimating the value of live beef cattle prior to slaughter. In this paper, we attempt to design a network structure to estimate the Beef Marbling Standard (BMS) from ultrasonic images of live beef cattle and individual information such as producer and father cattle. We demonstrated that our multi-input neural network with ultrasound images and individual information obtained a high accuracy in BMS estimation.