Investigate augmented vision technologies to improve subjective carcase assessment

Summary

Australia is one of the leading red meat producers in the world. The industry is keenly focussed on innovation and the delivery of new technology to drive efficiency and quality across the sector. This focus is in large part to ensure our future as a high value producer of red meat to the domestic and international markets.

Mixed reality is a term that spans the continuum between Augmented Reality and Virtual Reality. The global technology industry is pushing innovation in this area and bringing new and powerful solutions to market at great pace. Virtual Reality is targeted at the entertainment market but augmented reality is shaping up as a business and commercial set of tools. These tools have the capacity, as the name suggests, to augment the vision, knowledge, available information and tools that the user has at their disposal. This facilitates greater performance in terms of throughput and consistency of decision making by the user and is this area that is the focus of this project.
 
This project consists of two phases;A research phase reviewing and cataloguing relevant augmented reality projects, case studies and technologyA proof of concept phase in which a prototype of an augmented reality application is developed for an AR head mounted display and tasked with reducing the subjectivity in MSA grading in a processing environment.
 
The current meat grading practices have led to distrust in the industry. MLA have stated that producers and feedlot operators are concerned about the precision of meat grading in Australia. Meat graders are not to blame. Humans are simply not built to repeatedly make objective judgements day in day out. In an American study of meat grading, it was found that 50% of meat samples were mis-graded in some way. In a 2003 study of meat grading in Australia as many as 70% of samples were mis-graded (Jang Ju Won et al, 2017).

The solution is objective measurement in an easy to use package. With augmented reality, it is possible to integrate the objective measurement capability of computer vision with the experience of meat graders.  This will facilitate faster, more consistent and more precise meat grading while taking full advantage of the capabilities of meat graders.
The proof of concept developed is capable of directly discerning the colour of a meat sample, and count the area of the latissimus dorsi muscle and handsfree scanning of meat sample tickets. These features have been demonstrated on a Vuzix augmented reality headset as well as on various hand-held devices.  The major challenges to deployment on a head-mounted device are the lack of processing power and the heavy head-mounted device itself. 

Both the augmented reality market and the platform developed through this research show substantial potential. The successful demonstration of a meat grading application in conjunction with the continuing development of augmented reality solutions make it reasonable to expect augmented reality to play a substantial role in the meat industry in years to come.