A new Poultry CRC Project is exploring the potential of using machine vision technology in the cage layer industry.
Project Leader, Greg Cronin, a Research Scientist with DPI Victoria’s Animal Welfare Science Centre (AWSC), is aiming to determine whether machine vision can be used to count hens in cages and to identify blockages on the egg collection belt.
Machine vision technology, or automatic video image analysis, has been around for about 15 years and is used in many different industrial applications.
“If the project successfully demonstrates that machine vision can be used in the modern cage layer industry to count hens and monitor the egg collection belt, then it should be possible to broaden the applications to assist in the management of birds, through more frequent monitoring, for example, as well as the egg collection process,” says Greg.
The long-term objective of this proof-of-concept project is to identify work tasks that are repetitive, time consuming and able to be automated and monitored by computers.
“In the event that the computer detects a risk issue,” explains Greg, “then the stockperson would be instructed to investigate and resolve the problem.”
“Thus, stockpeople’s time could be spent in more productive activities, with low-risk, mundane activities completed automatically by the computer.”
The project also involves machine vision research engineer, Mark Dunn, from the National Centre for Engineering in Agriculture at the University of Southern Queensland, multimedia engineer, John McPherson, and Samantha Borg, a senior technical officer with DPI Victoria’s AWSC.
Greg, whose background is in agricultural science, animal behaviour and animal welfare, has been working with Poultry CRC Program Manager, John Barnett, on AECL-funded poultry welfare projects for a number of years.
He recently helped develop software designed to track a target attached to a pig through three-dimensional space and is currently involved in an AECL-funded project investigating the importance of nests for the welfare of layer hens.
“For the AECL project, I devised a method to record on video the egg-laying behaviour of hens in commercial cages and nest boxes so that we could record every egg laid, identify where and when each egg was laid and who laid it,” says Greg.
“This knowledge has been very useful for the Poultry CRC machine vision project and I’m very happy that the CRC agreed to fund it.”