Knowing the number of pigs on a large-scale pig farm is an important issue for efficient farm management. However, counting the number of pigs accurately is difficult for humans because pigs do not obediently stop or slow down for counting. In this study, we propose a camera-based automatic method to count the number of pigs passing through a counting zone. That is, using a camera in a hallway, our deep-learning-based video object detection and tracking method analyzes video streams and counts the number of pigs passing through the counting zone. Furthermore, to execute the counting method in real time on a low-cost embedded board, we consider the tradeoff between accuracy and execution time, which has not yet been reported for pig counting. Our experimental results on an NVIDIA Jetson Nano embedded board show that this “light-weight” method is effective for counting the passing-through pigs, in terms of both accuracy (i.e., 99.44%) and execution time (i.e., real-time execution), even when some pigs pass through the counting zone back and forth.

While the pig industry is crucial in global meat consumption, accounting for 34% of total consumption, respiratory diseases in pigs can cause substantial economic losses to pig farms. To alleviate this issue, we propose an advanced audio-visual monitoring system for the early detection of coughing, a key symptom of respiratory diseases in pigs, that will enhance disease management and animal welfare. The proposed system is structured into three key modules: the cough sound detection (CSD) module, which detects coughing sounds using audio data; the pig object detection (POD) module, which identifies individual pigs in video footage; and the coughing pig detection (CPD) module, which pinpoints which pigs are coughing among the detected pigs. These modules, using a multimodal approach, detect coughs from continuous audio streams amidst background noise and accurately pinpoint specific pens or individual pigs as the source. This method enables continuous 24/7 monitoring, leading to efficient action and reduced human labor stress. It achieved a substantial detection accuracy of 0.95 on practical data, validating its feasibility and applicability. The potential to enhance farm management and animal welfare is shown through proposed early disease detection.

As global pork consumption rises, livestock farms increasingly adopt deep learning-based automated monitoring systems for efficient pigsty management. Typically, a system applies a pre-trained model on a source domain to a target domain. However, real pigsty environments differ significantly from existing public datasets regarding lighting conditions, camera angles, and animal density. These discrepancies result in a substantial domain shift, leading to severe performance degradation. Additionally, due to variations in the structure of pigsties, pig breeds, and sizes across farms, it is practically challenging to develop a single generalized model that can be applied to all environments. Overcoming this limitation through large-scale labeling presents considerable burdens in terms of time and cost. To address the degradation issue, this study proposes a self-training-based domain adaptation method that utilizes a single label on target (SLOT) sample from the target domain, a genetic algorithm (GA)-based data augmentation search (DAS) designed explicitly for SLOT data to optimize the augmentation parameters, and a super-low-threshold strategy to include low-confidence-scored pseudo-labels during self-training. The proposed system consists of the following three modules: (1) data collection module; (2) preprocessing module that selects key frames and extracts SLOT data; and (3) domain-adaptive pig detection module that applies DAS to SLOT data to generate optimized augmented data, which are used to train the base model. Then, the trained base model is improved through self-training, where a super-low threshold is applied to filter pseudo-labels. The experimental results show that the proposed system significantly improved the average precision (AP) from 36.86 to 90.62 under domain shift conditions, which achieved a performance close to fully supervised learning while relying solely on SLOT data. The proposed system maintained a robust detection performance across various pig-farming environments and demonstrated stable performance under domain shift conditions, validating its feasibility for real-world applications.

Identifying poor-laying hens is important for reducing feed waste, but visual screening is labor-intensive. Motivated by the relationship between laying status and pigmentation, we propose CHIC (Color-Histogram Integrated Classifier), a lightweight model combining HVI (Horizontal, Vertical, Intensity) transformation with differentiable histogram encoding of learned feature activations and global-average-pooled features. Evaluation used a four-class head-and-leg patch dataset collected from 20 hens in six cages at a single facility, with cage-level laying-status labels. Across ten training runs with different development-data folds and a fixed frame-level test set, CHIC achieved mean binary F1-scores of 92.21% for head patches and 96.06% for leg patches, exceeding the MobileNetV4-S baselines. With 2.69 million parameters, the classifier had a mean forward-pass latency of 21.87 ms per patch on an NVIDIA Jetson Orin Nano Super. Component comparisons showed configuration-dependent effects; histogram integration also benefited grayscale inputs, while mean–standard-deviation pooling with HVI inputs achieved similar mean performance to CHIC. The classification advantage was maintained when predictions from detected head and leg regions were aggregated into hen-level decisions. However, performance declined substantially on unseen cages. These findings support CHIC as a candidate for automated poor-laying hen screening, with further validation required across independent hens and farms.