Project
Automated broiler phenotyping
The project aims to identify, upscale and implement technologies for automated individual broiler phenotyping to improve the currently recorded and novel phenotypic traits related to health, welfare and performance. Through extensive collaboration with broiler breeding practice and technological partners, we dedicate ourselves to research towards full practical implementation.
Broiler breeding
In broiler breeding, selection of the best animals for the next generation is of great importance to ensure that birds can adapt to future changes or demands of society and consumers.
For this, broiler breeding companies require individual records of the phenotypes of many birds.
However, manual or visual observations still form the basis of a considerable part of the phenotypic records that are routinely collected. These records can be time-consuming to collect and are mostly subjective. Furthermore, there is an interest in collecting data on novel traits, to obtain a more complete picture of broiler health, welfare and performance.
The Automated broiler phenotyping project
In the project, we will develop and implement an automated broiler phenotyping setup for recording activity, leg health and body weight at the individual level, while broilers are housed in groups. This encompasses:
- An automated activity tracking setup using radio frequency identification (RFID) technology and computer vision (CV)
- An automated leg health recording method using CV
- An automated body weight recording tool using RFID
- A combined approach to record activity, body weight and leg health simultaneously in the home pen of broilers.
The research will be performed in a real-life broiler breeding environment,
ensuring suitability for future practical implementation. Additionally, we will
estimate genetic parameters for broiler activity, leg health and body weight
and will examine whether activity levels can be used as a predictor or early
warning of declining health or welfare.
Selecting for improved health, welfare and performance
The results of this project will contribute to the development of generic and automated quantification methods for broiler health, welfare and performance, that are suitable for practical implementation. This will result in earlier-in-life and more objective health, welfare and performance scoring, and can subsequently contribute to faster genetic progress. Through implementation of the recordings in breeding programs, this project will contribute to improved health and robustness of broilers.
Papers
Conference contributions
-
Automated pose estimation reveals walking differences between normal and lame broilers
In: Proceedings of the ISAE Benelux conference 2022 - International Society for Applied Ethology (ISAE) -
Using radio frequency identification technology to track broiler activity: effects of different antenna configurations
-
Automated activity tracking in broilers and applications of the resulting individual-level data
In: Proceedings XI European Symposium on Poultry Welfare (ESPW 2023) - Prague: WPSA - ISBN: 9788090811416 - p. 36-36. -
Walking well or waddling: automated assessment of broiler leg health
-
Automatic phenotyping of walking ability in group-housed broilers using a multi-view camera setup
-
Computer vision derived pose features are associated with lameness in broilers
In: Proceedings XI European Symposium on Poultry Welfare (ESPW 2023) - Prague: WPSA - ISBN: 9788090811416 - p. 49-49.