Recent Applications in Computer Vision-based Lameness Detection of Dairy Cattles

Abstract

Lameness is one of the three widespread disorders leads economic losses due to reduced milk and early culling in dairy industry. It affects herd productivity and animal welfare and requires effective management strategies in order to minimize its impact. Therefore, it is essential to have precise and applicability detection of lameness in dairy farming, with a focus on environmental sustainability. The visual locomotion score approach to identification of lameness in lame cows based on qualitative analysis, involving the evaluation of different combinations of bodily abnormalities, it requires observation skills and specialization used in field. The use of computer vision (CV) technology in monitoring gait and posture of dairy cattle the collection of video data on timeline is particularly suitable for rapidly diagnostics. The processes involved in determining locomotion score in CV described as follows: data collection, preprocessing, identification of body regions, extraction of features and training of classifier. This research includes, the latest developments based on computer vision in the literature which used in research and application for the detection of lameness in dairy cattle.