Xiaoli Wei

Dry matter intake (DMI) in dairy cows is a critical indicator of nutrient intake from feed, serving as the cornerstone of precision feeding practices, playing a critical role in improving production efficiency and enhancing the quality of dairy products. To address the high costs of traditional measurement methods and the structural complexity and large parameter counts of neural network models, this study proposes a Stacking ensemble learning model to assess DMI, with model parameters optimized using the Tuna Swarm Optimization (TSO) algorithm to enhance assessment accuracy, taking cow body weight, lying duration, lying times, rumination duration, foraging duration, walking steps, and the concentrate-to-roughage feed ratio as input variables. To further improve TSO’s search efficiency and spatial exploration, this study introduces Sine–Logistic chaotic mapping, Levy flight, and Gaussian random walk strategy to optimize the TSO algorithm, developing the improved Tuna Swarm Optimization (ITSO). ITSO-optimized Stacking model achieved superior performance in DMI assessment, with an accuracy of 95.84%, significantly outperforming SVR, RF, DT, GBR, ETR, and AdaBoost models. This study provides a robust tool for precision feeding, contributing to optimizing cow feeding strategies, improving farm efficiency, and supporting sustainable dairy farming practices.

In real-world dairy farming environments, object recognition models often suffer from missed or false detections due to complex backgrounds and cow occlusions. In response to these issues, this paper proposes FSCA-YOLO, a multi-object cow behavior recognition model based on an improved YOLOv11 framework. First, the FEM-SCAM module is introduced along with the CoordAtt mechanism to enable the model to better focus on effective behavioral features of cows while suppressing irrelevant background information. Second, a small object detection head is added to enhance the model’s ability to recognize cow behaviors occurring at the distant regions of the camera’s field of view. Finally, the original loss function is replaced with the SIoU loss function to improve recognition accuracy and accelerate model convergence. Experimental results show that compared with mainstream object detection models, the improved YOLOv11 in this section demonstrates superior performance in terms of precision, recall, and mean average precision (mAP), achieving 95.7% precision, 92.1% recall, and 94.5% mAP—an improvement of 1.6%, 1.8%, and 2.1%, respectively, over the baseline YOLOv11 model. FSCA-YOLO can accurately extract cow features in real farming environments, providing a reliable vision-based solution for cow behavior recognition. To support specific behavior recognition and in-region counting needs in multi-object cow behavior recognition and tracking systems, OpenCV is integrated with the recognition model, enabling users to meet the diverse behavior identification requirements in groups of cows and improving the model’s adaptability and practical utility.

Dairy cow behavior carries important health information. Timely and accurate detection of behaviors such as drinking, feeding, lying, and standing is meaningful for monitoring individual cows and herd management. In this study, a model called Res-DenseYOLO is proposed for accurately detecting the individual behavior of dairy cows living in cowsheds. Specifically, a dense module was integrated into the backbone network of YOLOv5 to strengthen feature extraction for actual cowshed environments. A CoordAtt attention mechanism and SioU loss function were added to enhance feature learning and training convergence. Multi-scale detection heads were designed to improve small target detection. The model was trained and tested on 5516 images collected from monitoring videos of a dairy cowshed. The experimental results showed that the performance of Res-DenseYOLO proposed in this paper is better than that of Fast-RCNN, SSD, YOLOv4, YOLOv7, and other detection models in terms of precision, recall, and mAP metrics. Specifically, Res-DenseYOLO achieved 94.7% precision, 91.2% recall, and 96.3% mAP, outperforming the baseline YOLOv5 model by 0.7%, 4.2%, and 3.7%, respectively. This research developed a useful solution for real-time and accurate detection of dairy cow behaviors with video monitoring only, providing valuable behavioral data for animal welfare and production management.

The aim of this study is to identify an alternative approach for simulating the in vitro fermentation and quantifying the production of rumen methane and rumen acetic acid during the rumen fermentation process with different total mixed rations. In this experiment, dietary nutrient compositions (neutral detergent fiber (NDF), acid detergent fiber (ADF), crude protein (CP), and dry matter (DM)) were selected as input parameters to establish three prediction models for rumen fermentation parameters (methane and acetic acid): an artificial neural network model, a genetic algorithm-bp model, and a support vector machine model. The research findings show that the three models had similar simulation results that aligned with the measured data trends (R² ≥ 0.83). Additionally, the root mean square errors (RMSEs) were ≤1.85 mL/g in the rumen methane model and ≤2.248 mmol/L in the rumen acetic acid model. Finally, this study also demonstrates the models’ capacity for generalization through an independent verification experiment, as they effectively predicted outcomes even when significant trial factors were manipulated. These results suggest that machine learning-based in vitro rumen models can serve as a valuable tool for quantifying rumen fermentation parameters, guiding the optimization of dietary structures for dairy cows, rapidly screening methane-reducing feed options, and enhancing feeding efficiency.

Vision-based measuring method for individual cow feed intake using depth images and a Siamese network

The management of body condition score (BCS) during the dry period is associated with the postpartum health outcomes of dairy cows. However, the difference between the actual BCS and the fixed ideal value is not able to accurately predict the occurrence of postpartum diseases. This study aimed to use statistical process control (SPC) technology to monitor the BCS of dry cows, to evaluate the effect of control charts on nutritional strategies, and to explore the utility of SPC in predicting the incidence of postpartum subclinical ketosis (SCK). The BCS and SCK data of 286 cows from the dry off period to 60 days postpartum were collected to set up the early warning function. Three control charts, including a control chart for the average BCS of the herds, for the BCS of each dry cow, and for individual BCS, were established. The early warning signs for postpartum SCK development were: (1) an individual BCS more than 3.5 that remained unchanged for six weeks; (2) a capability index (CPK), an SPC tool, greater than −0.52. Using these parameters, the early warning signs of SCK development were verified in 429 dry cows. The results showed that the accuracy of early warning signal was 0.64 and the precision was 0.26. The control chart showed that the average BCS of dry cows was consistently higher than the expected upper limit of BCS during the experimental period, and that the addition of new cows to the herds increased the average BCS. In summary, the application of SPC technology to monitor the BCS of dry cows was not a good tool for the prediction of postpartum SCK occurrence but was an appropriate tool for guiding positive nutrition strategies.