Pig husbandry constitutes a significant segment within the broader framework of livestock farming, with porcine well-being emerging as a paramount concern due to its direct implications on pig breeding and production. An easily observable proxy for assessing the health of pigs lies in their daily patterns of movement. The daily movement patterns of pigs can be used as an indicator of their health, in which more active pigs are usually healthier than those who are not active, providing farmers with knowledge of identifying pigs’ health state before they become sick or their condition becomes life-threatening. However, the conventional means of estimating pig mobility largely rely on manual observations by farmers, which is impractical in the context of contemporary centralized and extensive pig farming operations. In response to these challenges, multi-object tracking and pig behavior methods are adopted to monitor pig health and welfare closely. Regrettably, these existing methods frequently fall short of providing precise and quantified measurements of movement distance, thereby yielding a rudimentary metric for assessing pig health. This paper proposes a novel approach that integrates optical flow and a multi-object tracking algorithm to more accurately gauge pig movement based on both qualitative and quantitative analyses of the shortcomings of solely relying on tracking algorithms. The optical flow records accurate movement between two consecutive frames and the multi-object tracking algorithm offers individual tracks for each pig. By combining optical flow and the tracking algorithm, our approach can accurately estimate each pig’s movement. Moreover, the incorporation of optical flow affords the capacity to discern partial movements, such as instances where only the pig’s head is in motion while the remainder of its body remains stationary. The experimental results show that the proposed method has superiority over the method of solely using tracking results, i.e., bounding boxes. The reason is that the movement calculated based on bounding boxes is easily affected by the size fluctuation while the optical flow data can avoid these drawbacks and even provide more fine-grained motion information. The virtues inherent in the proposed method culminate in the provision of more accurate and comprehensive information, thus enhancing the efficacy of decision-making and management processes within the realm of pig farming.

Identifying individual pigs is crucial for efficient breeding, health management, and disease control in modern farming. Traditional animal face identification methods are labor-intensive and prone to inaccuracies, while existing CNN-based pig face recognition models often struggle with high computational demands, large sizes, and reliance on extensive labeled data, which limit their practical application. This paper addresses these challenges by proposing a novel, decoupled approach to pig face recognition that separates detection from identification. This strategy employs a detection model as a pre-processing step, significantly reducing the need for extensive re-annotation for new datasets. Additionally, the paper introduces a method that integrates offline knowledge distillation with a lightweight pig face recognition model, aiming to build an efficient and embedding-friendly system. To achieve these objectives, the study constructs a small-scale, high-quality pig face detection dataset consisting of 1500 annotated images from a selection of 20 pigs. An independent detection model, trained on this dataset, then autonomously generates a large-scale pig face recognition dataset with 56 pig classes. In the face recognition stage, a robust teacher model guides the student model through a distillation process informed by a knowledge distillation loss, enabling the student model to learn relational features from the teacher. Experimental results confirm the high accuracy of the pig face detection model on the small-scale detection dataset and the ability to generate a large-scale dataset for pig face recognition on unlabeled data. The recognition experiments further verify that the distilled lightweight model outperforms its non-distilled counterparts and approaches the performance of the teacher model. This scalable, cost-effective solution shows significant promise for broader computer vision applications beyond agriculture.

Accurate animal face recognition is essential for effective health monitoring, behavior analysis, and productivity management in smart farming. However, environmental obstructions and animal behaviors complicate identification tasks. In pig farming, fences and frequent movements often occlude essential facial features, while high inter-class similarity makes distinguishing individuals even more challenging. To address these issues, we introduce the Pig Face Recognition and Inpainting System (PigFRIS). This integrated framework enhances recognition accuracy by removing occlusions and restoring missing facial features. PigFRIS employs state-of-the-art occlusion detection with the YOLOv11 segmentation model, a GAN-based inpainting reconstruction module using AOT-GAN, and a lightweight recognition module tailored for pig face classification. In doing so, our system detects occlusions, reconstructs obscured regions, and emphasizes key facial features, thereby improving overall performance. Experimental results validate the effectiveness of PigFRIS. For instance, YOLO11l achieves a recall of 94.92% and a ${\mathrm{AP}}_{50}$ of 96.28% for occlusion detection, AOTGAN records a FID of 51.48 and an SSIM of 91.50% for image restoration, and EfficientNet-B2 attains an accuracy of 91.62% with an F1 Score of 91.44% in classification. Additionally, heatmap analysis reveals that the system successfully focuses on relevant facial features rather than irrelevant occlusions, enhancing classification reliability. This work offers a novel and practical solution for animal face recognition in smart farming. It overcomes the limitations of existing methods and contributes to more effective livestock management and advancements in agricultural technology.

Pig vocalizations are important indicators of health and emotional state, offering significant potential for advancing precision livestock farming. Accurate recognition of these vocalization patterns requires robust noise filtering and effective feature extraction. However, existing studies often focus on isolated patterns such as coughs, limiting their practical applicability in real-world settings. This study introduces the Pig Vocalization Multi-stage Classification (PVMC) model, a comprehensive framework designed to detect and classify a wide range of pig vocalizations under diverse farm conditions for assessing health and emotional stress. PVMC adopts a multi-stage approach that integrates cough and scream detection with emotional state classification, providing a holistic analysis of pig vocalizations. The proposed system features: (1) improved robustness across varying vocalization durations and noise levels, (2) customized model architectures optimized for each stage of the pipeline, and (3) an ensemble learning strategy combining Wav2Vec2 and AST (Audio Spectrogram Transformer) to enhance performance and computational efficiency. PVMC achieved a signal-to-noise ratio (SNR) improvement of up to 4.9dB, 95.80% accuracy in vocalization segmentation, 98.88% accuracy in key vocalization classification, and 92.15% accuracy in emotional state detection. Notably, the ensemble method significantly improved overall precision, recall, and F1-score. These results demonstrate the PVMC model’s robustness and practical utility as a deployable solution for real-time pig vocalization monitoring, contributing to intelligent, welfare-oriented livestock management systems.

Environmental air anomaly detection is crucial for ensuring the healthy growth of livestock in smart pig farming systems. This study focuses on four key environmental variables within pig housing: temperature, relative humidity, carbon dioxide concentration, and ammonia concentration. Based on these variables, it proposes a novel encoder–decoder architecture for anomaly detection based on continuous-time models. The proposed framework consists of two embedding layers: an encoder module built around a continuous-time neural network, and a decoder composed of multilayer perceptrons. The model is trained in a self-supervised manner and optimized using a reconstruction-based loss function. Extensive experiments are conducted on a multivariate multi-sequence dataset collected from real-world pig farming environments. Experimental results show that the proposed architecture significantly outperforms existing transformer-based methods, achieving 92.39% accuracy, 92.08% precision, 85.84% recall, and an F₁ score of 88.19%. These findings highlight the practical value of accurate anomaly detection in smart farming systems; timely identification of environmental irregularities enables proactive intervention, reduces animal stress, minimizes disease risk, and ultimately improves the sustainability and productivity of livestock operations.