Artificial intelligence application in carcass beef grading automation

Abstract

The carcass grading process can be divided into two subprocesses: gathering of relevant information through visual observations and decision-making based on the relevant information to assign the final yield and quality grades. Both subprocesses require expert knowledge in terms of what information is needed and how to use set rules for deducing the final grades. An automated meat grading system simulates the observational and reasoning skills of a human meat grader using a computer vision system operated in a cluttered, relatively unconstrained environment. The automated meat grading system consists of a computer vision subsystem, which obtains the carcass characteristics, and a knowledge-based meat grading subsystem, which uses this information and the set rules to determine the grades for results. This paper presents the technical aspects of developing an automated carcass beef grading system at the US Meat Animal Research Center at Clay Center, NE.<>