A two-stage recognition method based on deep learning for sheep behavior
Published in 2023
Scientific article on a method to detect sheep behavior based on deep learning. It recognizes six types of sheep behavior, three corresponding to normal physiological activities (standing, feeding and lying down) and three behaviors considered to be disruptive (attacking, biting and climbing) that require immediate action from the farmer.
Document Types: Scientific paper
Animal categories: Ovines
Keywords:Animal-based measurements, Precision farming, Welfare indicators, Human-animal relationships