Thousands of cattle pass through the Casino Food Co-op daily. For the northern NSW business, ensuring the well-being of the cattle is not just business, its law. Currently, the cattle are monitored by staff and vets, but the task of monitoring every animal is extremely difficult. But new AI technology is being developed to change that.
The question of how much the AI can cover was a prominent one. Could it monitor individual animals to ensure their welfare? Could it know when an issue requires action from management? Could AI ensure that animals have all they require in the way of food, water, and habitat twenty-four seven? A new program can currently detect when individual cows are standing, walking or lying down.
CSIRO’s animal welfare and behaviour analysts and their data specialists have teamed up with Casino Food Co-op. Together, they are in the process of developing an AI powered monitoring system to keep an eye on livestock.
Monitoring animal welfare is a difficult task, particularly when cattle are unloaded into yards after road transport. Many yards house thousands of cattle each. For highly trained vets and staff, round-the-clock monitoring of every animal is a wildly unrealistic goal. Scientists are hoping that an AI-equipped camera system will be able to fill this role and keep a constant eye on the cattle.
Programming the AI to identify cattle behaviours correctly is difficult, as cattle of the same breed can look very similar. Cattle have similar shapes, sizes, colours, and a tendency to stick together in a herd. As such, the researchers are having to develop software that can, via a camera system, automatically identify cattle behaviours. The software helps figure out whether a cow herd is relaxed, hydrated, or showing signs of injury.
The program helps support on-site vets to identify animals in need of help or at risk, making it possible to take more timely action to help. The AI program also assists management teams on how to improve their performance and ensures compliance with the current industry standards for animal welfare.
The program was designed to prove its usefulness for the industry in real world conditions, due to law changes happening in 2026. From 2026, video surveillance will be mandatory for Australian Meat Industry Council certificated processors under the Australian Livestock Processing Industry Animal Welfare Certification System. “AI-powered monitoring could be a valuable addition to existing methods for ensuring animal welfare,” said Dr Caroline Lee, a Senior Principal Research Scientist with CSIRO.
“Automating inspections will help the industry improve animal welfare,” added Dr Dadong Wang, Research Group Leader.” The technologies CSIRO is developing will help make sense of the videos captured at the processing facility, provide objective evidence of animal welfare assurance, and support better decision making for management,” he said. There are plans currently being made by the team to expand from tracking animals in holding pens to monitoring unloading ramps to improve early detection of issues.