Accurate cattle body detection is crucial for precision livestock farming. However, traditional cattle body detection methods rely on manual observation, which is both time-consuming and labor-intensive. Moreover, computer-vision-based methods suffer prolonged training times and training difficulties. To address these issues, this paper proposes a novel YOLOv5-EMA model for accurate cattle body detection. By incorporating the Efficient Multi-Scale Attention (EMA) module into the backbone of YOLO series detection models, the performance of detecting smaller targets, such as heads and legs, has been significantly improved. The Efficient Multi-Scale Attention (EMA) module utilizes the large receptive fields of parallel sub-networks to gather multi-scale spatial information and establishes mutual dependencies between different spatial positions, enabling cross-spatial learning. This enhancement empowers the model to gather and integrate more comprehensive feature information, thereby improving the effectiveness of cattle body detection. The experimental results confirm the good performance of the YOLOv5-EMA model, showcasing promising results across all quantitative evaluation metrics, qualitative detection findings, and visualized Grad-CAM heatmaps. To be specific, the YOLOv5-EMA model achieves an average precision (mAP@0.5) of 95.1% in cattle body detection, 94.8% in individual cattle detection, 94.8% in leg detection, and 95.5% in head detection. Moreover, this model facilitates the efficient and precise detection of individual cattle and essential body parts in complex scenarios, especially when dealing with small targets and occlusions, significantly advancing the field of precision livestock farming.

The efficient detection and counting of pig populations is critical for the promotion of intelligent breeding. Traditional methods for pig detection and counting mainly rely on manual labor, which is either time-consuming and inefficient or lacks sufficient detection accuracy. To address these issues, a novel model for pig detection and counting based on YOLOv5 enhanced with shuffle attention (SA) and Focal-CIoU (FC) is proposed in this paper, which we call YOLOv5-SA-FC. The SA attention module in this model enables multi-channel information fusion with almost no additional parameters, enhancing the richness and robustness of feature extraction. Furthermore, the Focal-CIoU localization loss helps to reduce the impact of sample imbalance on the detection results, improving the overall performance of the model. From the experimental results, the proposed YOLOv5-SA-FC model achieved a mean average precision (mAP) and count accuracy of 93.8% and 95.6%, outperforming other methods in terms of pig detection and counting by 10.2% and 15.8%, respectively. These findings verify the effectiveness of the proposed YOLOv5-SA-FC model for pig population detection and counting in the context of intelligent pig breeding.

Novel method for the recognition of Jinnan cattle action using bottleneck attention enhanced two-stream neural network