Pig weight and body size are important indicators for producers. Due to the increasing scale of pig farms, it is increasingly difficult for farmers to quickly and automatically obtain pig weight and body size. Due to this problem, we focused on a multiple output regression convolutional neural network (CNN) to estimate pig weight and body size. DenseNet201, ResNet152 V2, Xception and MobileNet V2 were modified into multiple output regression CNNs and trained on modeling data. By comparing the estimated performance of each model on test data, modified Xception was selected as the optimal estimation model. Based on pig height, body shape, and contour, the mean absolute error (MAE) of the model to estimate body weight (BW), shoulder width (SW), shoulder height (SH), hip width (HW), hip width (HH), and body length (BL) were 1.16 kg, 0.33 cm, 1.23 cm, 0.38 cm, 0.66 cm, and 0.75 cm, respectively. The coefficient of determination (R²) value between the estimated and measured results was in the range of 0.9879–0.9973. Combined with the LabVIEW software development platform, this method can estimate pig weight and body size accurately, quickly, and automatically. This work contributes to the automatic management of pig farms.

The accurate and rapid detection of objects in videos facilitates the identification of abnormal behaviors in pigs and the introduction of preventive measures to reduce morbidity. In addition, accurate and effective pig detection algorithms provide a basis for pig behavior analysis and management decision-making. Monitoring the posture of pigs can enable the detection of the precursors of pig diseases in a timely manner and identify factors that impact pigs’ health, which helps to evaluate their health status and comfort. Excessive sitting represents abnormal behavior when pigs are frustrated in a restricted environment. The present study focuses on the automatic recognition of standing posture and lying posture in grouped pigs, which shows a lack of recognition of sitting posture. The main contributions of this paper are as follows: A human-annotated dataset of standing, lying, and sitting postures captured by 2D cameras during the day and night in a pig barn was established, and a simplified copy, paste, and label smoothing strategy was applied to solve the problem of class imbalance caused by the lack of sitting postures among pigs in the dataset. The improved YOLOX has an average precision with an intersection over union threshold of 0.5 (AP_0.5) of 99.5% and average precision with an intersection over union threshold of 0.5–0.95 (AP_0.5–0.95) of 91% in pig position detection; an AP_0.5 of 90.9% and an AP_0.5–0.95 of 82.8% in sitting posture recognition; a mean average precision with intersection over union threshold of 0.5 (mAP_0.5) of 95.7% and a mean average precision with intersection over union threshold of 0.5–0.95 (mAP_0.5–0.95) of 87.2% in all posture recognition. The method proposed in our study can improve the position detection and posture recognition of grouped pigs effectively, especially for pig sitting posture recognition, and can meet the needs of practical application in pig farms.

Aggressive behavior among pigs is a significant social issue that has severe repercussions on both the profitability and welfare of pig farms. Due to the complexity of aggression, recognizing it requires the consideration of both spatial and temporal features. To address this problem, we proposed an efficient method that utilizes the temporal shift module (TSM) for automatic recognition of pig aggression. In general, TSM is inserted into four 2D convolutional neural network models, including ResNet50, ResNeXt50, DenseNet201, and ConvNext-t, enabling the models to process both spatial and temporal features without increasing the model parameters and computational complexity. The proposed method was evaluated on the dataset established in this study, and the results indicate that the ResNeXt50-T (TSM inserted into ResNeXt50) model achieved the best balance between recognition accuracy and model parameters. On the test set, the ResNeXt50-T model achieved accuracy, recall, precision, F1 score, speed, and model parameters of 95.69%, 95.25%, 96.07%, 95.65%, 29 ms, and 22.98 M, respectively. These results show that the proposed method can effectively improve the accuracy of recognizing pig aggressive behavior and provide a reference for behavior recognition in actual scenarios of smart livestock farming.

Egg production rate and egg weight are core indicators for evaluating the production performance of broiler breeders. The accurate prediction of these indicators can significantly enhance farm economic efficiency and can provide a basis for future production strategies. Currently, there is a lack of research on the application of machine learning (ML) models to predict egg production rate and egg weight in broiler breeders. In this study, we collected data on age, feed intake, water consumption, and environmental factors (temperature, humidity and wind speed) from three poultry houses to train the predictive models. Based on this data, we developed three different datasets. In each dataset, data from a single poultry house were divided into a training set and a validation set in an 8:2 ratio, and data from the remaining two poultry houses were combined to form the test set. We systematically compared the performances of the following seven ML models in predicting egg production rate and egg weight: random forest (RF), multilayer perceptron (MLP), support vector regression (SVR), least squares support vector machine (LSSVM), k-nearest neighbors (kNN), XGBoost, and LightGBM. The results indicated that the XGBoost model demonstrated the best performance across all three datasets. In predicting egg production rate, the XGBoost model achieved a mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE) of less than 2.86%, 4.17% and 7.03%, respectively. For egg weight predictions, the XGBoost model's MAE, RMSE and MAPE were less than 0.63g, 0.86g and 1.1%, respectively. Given the inherent black-box nature of ML models, we used the Shapley additive explanations (SHAP) method to interpret the key features influencing the XGBoost model's predictions and the interactions between these features. The key features for predicting egg production rate are age, feed intake and effective temperature (ET). For egg weight prediction, the most important features are age, wind speed, temperature-humidity index (THI) and ET. This approach enhanced the model's transparency and credibility. This study provides scientific evidence for predicting the production performance of broiler breeders. Accurately predicting egg production rate and egg weight provides a scientific basis for farm operations, aiding in optimizing resource allocation, improving production efficiency, enhancing animal welfare, and ultimately boosting the farm's profitability.

Feeding and drinking behaviors are important in pig breeding. Although many methods have been developed to monitor them, most are too expensive for pig research, and some vision-based methods have not been integrated into equipment or systems. In this study, two systems were designed to monitor pigs’ feeding and drinking behaviors, which could reduce the impact of the image background. Moreover, three convolutional neural network (CNN) algorithms, VGG19, Xception, and MobileNetV2, were used to build recognition models for feeding and drinking behaviors. The models trained by MobileNetV2 had the best performance, with the recall rate higher than 97% in recognizing pigs, and low mean square error (RMSE) and mean absolute error (MAE) in estimating feeding (RMSE = 0.58 s, MAE = 0.21 s) and drinking durations (RMSE = 0.60 s, MAE = 0.12 s). In addition, the two best models trained by MobileNetV2 were combined with the LabVIEW software development platform, and a new software to monitor the feeding and drinking behaviors of pigs was built that can automatically recognize pigs and estimate their feeding and drinking durations. The system designed in this study can be applied to behavioral recognition in pig production.

In recent years, the poultry breeding industry has been converted into a large-scale, intensive, and intelligent production mode. The Internet of Things (IoT) is under rapid development, which promotes the development of precision livestock farming. In this study, we developed an intelligent service platform for a facility environment based on the IoT structure, utilizing the capabilities of Platform as a Service (PaaS). The platform consists of four layers, including an information perception layer, network layer, management service layer, and application layer. By using the cloud service model with a distributed network architecture, asynchronous data transmission, and a distributed file system, the platform can centrally manage multiple farm’s data. The intelligent service platform includes the following functions: displaying environmental data, water and electricity consumption, data analysis, and managing production data. Over a 500-day trial period in a live poultry house, the platform demonstrated high data integrity (>87%) and resilience against network disruptions and power outages. The data validity of each environmental element exceeded 94%, among which the validity of the temperature, humidity, and carbon dioxide concentration exceeded 99%. The overall accuracy of the dataset remained relatively high, providing a robust data foundation for further research. Key features included audio analysis, environmental monitoring, and production data management. The platform’s operational status was efficiently communicated via data statistics and email alerts, facilitating timely system recovery. The demonstrated modules included sound recognition, psychrometric charts for visual alerts, and financial analysis tools, offering versatile solutions for integrating PLF models and advanced analytics.