Automated, real-time monitoring of bird movement and behaviour

This project improves the welfare and mortality rates of birds in commercial pens using AI-based video analytics technologies to track bird movement and pen density.

Artificial intelligence

Data science

}

Duration

18 months

Academic team

Associate Professor Jian Zhang

Engagement model

Funding

Australian Eggs

Future applications

The platform technology can be applied to monitoring hen weight, behaviour, feeding etc. The technology can also be applied to other farm animals.

Challenge

The farming industry currently has no research, predictors of risk or controls in place to reduce pile-ups in commercial egg farms. A UK study found that an average of 25.5 birds are lost to smothering per production cycle as there are no warnings in place to alert staff to pile-ups.

Australian Eggs want a reliable, real-time solution to monitor pen density in order to improve hen welfare and reduce the risk of death.

 

Solution

The UTS Tech Lab team developed an AI-based video technology that monitors the real-time movement of birds in commercial sheds.

The system uses two cameras inside and outside the shed to detect individual bird movement. Using advanced computer vision and pattern recognition algorithms, the video-based system can monitor and report on movement in real-time.

The system will support farm management by offering warning signals when pen density is reaching or at capacity. The solution will also reduce labour requirements, minimise human intervention in sheds and reduce costs.

This technology platform can be applied to behaviour (e.g. pen bullying), feeding, drinking and animal counting of birds or other animals.