Objectif
The welfare of animals kept for human consumption is of high concern globally, also within the European Commission including the welfare of the domesticated pig . As animal well-being is not just the absence of negative affective states, but also, and probably predominantly, the presence of positive affective states, identifying indicators of positive affective states in farmed animals can increase our capacity to improve their health and welfare. One such indicator is play behaviour. Play in pigs is an event occurring sporadically. Thus, using play behaviour for welfare assessment will not be possible through direct observations. Instead, it demands automation and continuous monitoring, two key elements of Precision Livestock Farming (PLF). The current project aims to investigate the relation between play behaviour in pigs and sensor data from image and sound analysis combining ethology and computer science into one field of Computational Ethology (CE). The project will focus on the CE part of the research, but the knowledge obtained could be used to develop real-time PLF algorithms for use in pig herds to observe the frequency and duration of play behaviour at pen level, beneficial to both the farmer, consultants, inspectors and future research projects within play behaviour in pigs. Training of the reseacher will include a thorough introduction to the field of CE and PLF, to the decoding and labeling of feature variables and to the use of machine learning algorithms on sensor data; all expertise fields of the hosting research group. All in all, the current project will take the first steps in developing a method for automatic recognition of play behaviour and positive affective states in pigs, as well as training the researcher for a future career within Computational Ethology and Precision Livestock Farming.
Champ scientifique
Programme(s)
Régime de financement
MSCA-IF-EF-ST - Standard EFCoordinateur
3000 Leuven
Belgique