Analysis of images and sounds as indicators of welfare and health of poultry
L'analyse d'image et sonore pour le relevé d'indicateurs de bien-être et de santé des volailles
Résumé
Guaranteeing consumers that poultry farming respects animal welfare is the heart of the poultry farmer,
but civil society is demanding greater transparency in farming practices. Meeting society's expectations
must go hand in hand with the competitiveness of poultry meat production, which is globalised and highly
competitive. The notion of animal health and welfare can be assessed using a variety of methods, and
new technologies offer an opportunity to take continuous measurements in real time, without disturbing
the animals in their living environment. Image and sound processing enables finer and more frequent
analyses than those carried out by humans. These new technologies are helping to improve monitoring
and responsiveness to health problems or changes in animal behaviour through predictive analysis. The
EBroilerTrack project led by ITAVI has produced promising proofs of concept in imaging and acoustics
under controlled broiler rearing conditions. In the field of imaging, image analysis algorithms have been
developed to monitor individual broiler chickens on the farm, with the aim of identifying welfare and health
indicators for each animal observed. The performance of these algorithms is presented in this article. In
the field of acoustics, the work carried out has demonstrated the benefits of using acoustic analysis to
monitor the health and well-being of broilers, in the specific case of infectious bronchitis.
Origine | Fichiers éditeurs autorisés sur une archive ouverte |
---|---|
Licence |