Frontmatec Automated MSA/AUS-MEAT hyperspectral handheld grading for beef

Summary

Visual grading of carcases conducted by AUS-MEAT accredited graders has served the industry well for many years. Although vision-based systems for grading beef rib eye exist in the market, the commercially available equipment includes a number of limitations for Australian beef grading. Recent advances in handheld technology provides the opportunity for improved instrument-based objective measurement.

This project delivered an automated handheld camera that grades the beef rib eye according to MSA and AUS-MEAT standards consistently and in accordance with visual grading conducted by highly trained graders.

The precision and accuracy achieved against AUS-MEAT accreditation standards with the automated handheld camera has provided the technology company, Frontmatec, with the confidence to develop a commercial device.

Objectives

The overall project goal was to make an instrument available that has prediction models that will enable all abattoirs in Australia to grade carcases in a uniform and consistent way, communicate accurate supply chain feedback and thereby create significant value for the Australian beef industry.

Key findings

• The handheld rib eye camera developed by Frontmatec determines rib-eye area, marbling score and chemical intramuscular fat with required precision.
• Algorithms for ribeye area, intramuscular fat content, marbling score against both MSA and AUS-MEAT classification and meat colour have been developed.
• Subcutaneous rib fat thickness cannot be determined by the camera solution, as the line separating areas of intermuscular fat is not clear with machine imaging. An alternative to subcutaneous rib fat would be total fat thickness.
• Based on the learnings from the current project, Frontmatec intends to progress to developing a commercial device.

Benefits to industry

This project has developed an objective measurement device that will ensure consistent rib-eye grading results that provide the most accurate supply chain feedback.

This innovation will also eliminate any perceived or potential subjective influence through the development of this class of automated grading equipment.

MLA action

MLA will continue to support the next steps in development and widespread adoption of this technology. Interested readers are directed to request the ALMTech Traits Manual (currently in draft) for a comprehensive background on key traits, associated measurement approaches, and related metrology (ie; the science around measurement). Please contact reports@mla.com.au for further information.

Future research

Development of an algorithm for fat colour classification awaits a data set that covers a larger spread in colour classes with each class properly represented.