Vision-Based Cow Tracking and Feeding Monitoring for Autonomous Livestock Farming

Animal tracking and feeding monitoring is crucial for automatic individual cow welfare measurement and natu- rally becomes a prerequisite for autonomous livestock farming systems. The deformable body posture and irregular movement of cows under complex farming environment make tacking of individual animals in a herd very challenging. To improve the performance of face-based cow tracking and feeding monitoring, a deep learning network based approach, namely, YOLOv5s- CA+DeepSORT-ViT, was proposed in this paper. In our proposed approach, Coordinate Attention (CA) integrated YOLOv5 was developed to capture spatial location information to improve the face detection performance for overlapping regions. Then the Vision Transformer (ViT) was embedded in the re-identification network DeepSORT to enhance feature matching and tracking accuracy. The comparative results of the multi-cow complex dataset constructed from a commercial farm show that the ID F1 Score (IDF1) and Multi-target tracking accuracy (MOTA) of the proposed YOLOv5s-CA+DeepSORT-ViT are 88.5% and 84.4% respectively. Meanwhile, the ID switching (ID Sw.) times and the processing time are reduced by 50% and 20% compared to the YOLOv5s+DeepSORT model. Experimental results also showed that the overall cow tracking performance of our proposed ap- proach outperformed the other baselines (e.g. SORT, ByteTrack, BoT-SORT and DeepSORT).