Australian Eggs: UTS Tech Lab’s AI revolution in bird welfare

A real-time monitoring system, powered by fixed-in-place video cameras and sophisticated AI algorithms, is reshaping the way we approach bird welfare assessment on Australian farms.

Building upon the success of the first phase of the project, which monitored pen movement and density, the project’s next phase delves into comprehensive measurements, welfare assessments, and analyses of feeding, drinking patterns, and egg laying.

This rich dataset gives farmers valuable information for better decision-making, ultimately leading to improved flock welfare and performance.


The Australian egg industry confronts a considerable hurdle in precisely evaluating bird welfare on egg farms. Conventional methods can be impacted by observer biases and inflict stress on animals through frequent sampling. Furthermore, current management systems lack universality, resulting in disparate measurements and assessments.

A pressing requirement emerges for a solution that furnishes non-intrusive, repeatable, and universally consistent measurement systems, all while diminishing dependence on labour and mitigating stress on the animals.

The solution must also deliver a cost-effective and easily deployable platform for the efficient monitoring of flock behaviours.


UTS Tech Lab proposes an innovative solution leveraging advanced AI-based platform technology to revolutionise bird welfare measurement and assessment on egg farms.

The pioneering approach involves the development of a real-time monitoring system utilising strategically placed fixed video cameras. These cameras observe and analyse bird and flock behaviours objectively and consistently, setting a new standard for precision in welfare assessment.

This technology employs computer vision and machine learning techniques to create a large, annotated bird behaviour database, enabling the construction of accurate video algorithms for quantitative analysis.

  • Objective Measurement: The system delivers a bird behaviour census, ensuring accurate and real-time measurement of flock activities. By eliminating observer biases, this approach provides a reliable foundation for data collection.
  • Welfare Status Assessment: In collaboration with bird welfare experts, the system assesses positive and negative states associated with observed activities. This empowers farmers to gauge the welfare status of the flock, enabling informed decisions for interventions and enhancements.
  • Production-Related Monitoring: Simultaneously, the system monitors production-related behaviours such as feeding, drinking, and egg laying. This holistic approach allows for a comprehensive analysis of both welfare and production aspects, providing a complete view of flock performance.
  • Risk Factor Identification: The system’s management software utilizes data-driven insights to identify risk factors impacting welfare and production. By analysing correlations between measurements and inputs, farmers gain valuable information for better decision-making, ultimately leading to improved flock welfare and performance.

This innovative fusion of automated identification, behavioural analysis, and data-driven insights not only addresses the challenges faced by the Australian egg industry but also establishes a platform technology for modern farming practices.

2 years

Academic team
Associate Professor Jian Zhang
Associate Professor Qiang Wu
Dr Richard Shephard

Multimedia Data Analytics

Engagement model
Research as a service

Australian Eggs

Future applications
The technology can be applied to other farm animals.

Area of expertise
Artificial intelligence
Data science