Body condition is an important welfare measure in many livestock species, and under genetic control. Advances in computer vision and artificial intelligence (AI) offer the potential for automated, cost-effective phenotyping and continuous monitoring of traits like body condition. However, the use of computer vision and AI for phenotyping body condition score remains limited in commercial herds. We have collected more than 5 000 visual inspected phenotypes in Holstein and Jersey cattle along with camera data that can be used for prediction of body condition score for Danish herds equipped with the CFIT system.
This work focuses on developing machine learning-based predictive models using 3D imaging and assessing the potential of predicted BCS as a breeding trait through variance component estimation and breeding value prediction.
Jan Lassen, Senior project scientist
Viking Genetics
Email: jalas@vikinggenetics.com
Phone +45 20407441
To be decided in agreement with the supervisor.
We provide a study place at a student office in QGG Aarhus main campus or VikingGenetics, Assentoft, Randers
60 ECTS: Experimental theses in which the student is responsible for planning, trial design and collection and analysis of his/her own original data
Programming skills in Python and R is an advantage but not a prerequisite.
