THE University of Technology Sydney (UTS) has developed a portable tool using 3D imaging and artificial intelligence to provide rapid real-time estimation of key traits in live cattle and carcases to improve decision making.
An animal deposits muscle and fat in different areas of the body which results in different body shapes (phenotypes).
Typically trained human assessors visually inspect an animal and estimate key traits; however, such estimates are subjective, have low accuracy and reproducibility.
UTS has worked with the NSW Department of Primary Industries (DPI) and Meat & Livestock Australia (MLA) to develop a device for estimating P8 fat (mm), muscle and condition score using automated trait assessment.
The technology uses an off-the-shelf RGB-D camera to take 3D images of an animal as it passes through a race.
Benefits of the technology include:
- Live animal and carcase assessments
- BeefSpecs compliant for consistent classing and market compliance
- Accurate stock assessment (muscle score, fat content and hip height), improved stock valuations and reduced transaction risks
- Make better decisions and improve production by reducing input costs and variability in herd
- In conjunction with BeefSpecs, allow to optimise timing for sale
- In conjunction with BeefSpecs, tailor drafts for different finishing regimes based on predicted performance and market target
- Assess carcase information and aid management or marketing decisions
- Safe — the operator does not touch the animal
- Easy to set up and maintain, grading is performed within seconds
Applications include use by farmers/producers, saleyards/auctions, feedlots and abattoirs.
A patent application has been filed for this technology and we are seeking partners to work with our researchers to develop it further.