Search

• Spatiotemporal distribution of livestock behavior was monitored using multi-modal data. • Multi-source data fusion outperformed single-source data in multi-behavior classification. • The CNN-LSTM model was able to classify multi-behavior with higher accuracy. • 1HZ of multi-source fusion data still achieved reasonable classification accuracy. Monitoring multiple behavioral patterns and grazing trajectories of individual sheep can provide valuable information for various aspects of livestock production. The emergence of low-cost miniature sensors, coupled with the continuous advancement of deep learning technologies, has ushered in a new generation of intelligent solutions for precision livestock farming. This study aims to explore the deployment methods of motion sensors, selection of data collection frequencies, and choice of deep learning algorithms to provide accurate classification of multiple behaviors in grazing sheep. Based on this, in conjunction with the acquired location information, the goal is to comprehend the spatiotemporal distribution of grazing sheep behaviors. Devices capable of collecting Inertial Measurement Unit (IMU) and location data were attached to the jaw, neck, and hind leg of sheep. Four datasets were created using IMU data with a frequency of 20 Hz and a 5 s time window from different positions (neck, neck & leg, jaw, jaw & leg). Two deep learning models, Convolutional Neural Network (CNN) - Long Short Term Memory (LSTM) and Temporal Convolutional Network (TCN)-Transformer, were employed to classify six grazing sheep behaviors: walking, standing, grazing, lying, standing-ruminating, and lying-ruminating. The results indicate that by fusing data from the neck- and leg-mounted devices and utilizing the CNN-LSTM model, the accuracy reached the highest at 99.3 %. Furthermore, a comparison was made regarding the behavior classification accuracy of this fused data at different IMU data frequencies (20, 10, 5, and 1 Hz). It was found that even when the data frequency was reduced to 1 Hz, the classification accuracy for the six sheep behaviors still exceeded 96 %. Additionally, the trained model with the highest accuracy was applied to monitoring grazing sheep behavior under two different management procedures. Further analysis of the time budgets of sheep behavior revealed differences in the durations of behaviors under different management procedures. For extensively grazing sheep, location information was also monitored and combined with behavior classification results to generate the spatiotemporal distribution of sheep behavior. The technologies presented are important for gaining further insights into the health status of grazing livestock and grassland, thereby enhancing grazing management practices by informing grazing choices and optimizing grassland utilization. [ABSTRACT FROM AUTHOR]

In this paper, we consider the problem of multi-class image classification when the classes behaviour has a low rank structure. That is, classes can be embedded into a low dimensional space. Traditional multi-class classification algorithms usually use nuclear norm to approximate the rank of the weight matrix. Considering the limited ability of the nuclear norm for the accurate approximation, we propose a new scalable large scale multi-class classification algorithm by using the recently proposed truncated nuclear norm as a better surrogate of the rank operator of matrices along with multinomial logisitic loss. To solve the non-convex and non-smooth optimization problem, we further develop an efficient iterative procedure. In each iteration, by lifting the non-smooth convex subproblem into an infinite dimensional ℓ 1 norm regularized problem, a simple and efficient accelerated coordinate descent algorithm is applied to find the optimal solution. We conduct a series of evaluations on several public large scale image datasets, where the experimental results show the encouraging improvement of classification accuracy of the proposed algorithm in comparison with the state-of-the-art multi-class classification algorithms. [ABSTRACT FROM AUTHOR]

Abstract: An effective lubrication can significantly reduce wear of metal-on-metal artificial hip joints. The improvement of the lubrication can be achieved through the optimisation of the bearing geometry in terms of a small clearance and/or the structural support such as a polyethylene backing underneath a metallic bearing in a sandwich acetabular cup form. The separate effects of these two factors on fluid film lubrication of 28mm diameter metal-on-metal total hip joints under walking conditions were numerically investigated in this paper. The results show that a larger lubricant film due to the polyethylene backing can be significantly enhanced by the transient squeeze-film action, particularly during the stance phase, and a similar lubricant film can be developed for both the monolithic cup relying on the smaller clearance and the sandwich cup benefiting from the polyethylene backing. Both cup systems can function in a wide range of lubrication regimes, covering both mixed and fluid film, under the current design and manufacture conditions. [Copyright &y& Elsevier]

The Person Re-identification (ReID) task aims to match persons across cameras in a surveillance system. In the past few years, many researches are devoted to ReID and its performance has gained significant improvement. ReID models are usually trained as a joint framework comprising a person feature extractor and a classifier. However, there exists co-adaptation between the feature extractor and the classifier, which prevents the feature extractor from making effective and sufficient optimization and results in inferior retrieval performance. In this paper, we propose a very simple and effective training method, called DeAda, to decouple this co-adaptation. Our main motivation is to construct a series of weak classifiers during training by randomization of parameters, so that optimization on the feature extractor could be strengthened in the training stage. DeAda is easy, effective, and efficient, and could serve as a plug-and-play optimization tool for ReID models, without additional memory and time cost. We also analyze the theoretical property of DeAda and show that it could produce identical features for the same person under some simple assumptions. We demonstrate its effectiveness on three public ReID datasets: Market1501, DukeMTMC-reID and CUHK03 over different ReID models. With DeAda optimization, we finally obtain state-of-the-art results on all the three datasets. [ABSTRACT FROM AUTHOR]

• Heat stress modelling was done on an actual dairy farm equipped with sprinklers. • ANN models were the best to predict physiological responses in dairy cows. • Machine learning and its interpretation are relevant in heat stress research. • Individual heat stress states can be assessed non-invasively. Heat stress is increasingly affecting the production, health, and reproduction of dairy cows. Previous studies used limited variables as predictors of physiological responses, and the developed models poorly predict animal responses in evaporatively cooled environments. The aim of this study was to build machine learning models using comprehensive variables to predict physiological responses of dairy cows raised on an actual dairy farm equipped with sprinklers. Four algorithms including random forests, gradient boosting machines, artificial neural networks (ANN), and regularized linear regression were used to predict respiration rate (RR), vaginal temperature (VT), and eye temperature (ET) with 13 predictor variables from three dimensions: production, cow-related, and environmental factors. The classification performance of the predicted values in recognizing individual heat stress states was compared with commonly used thermal indices. The performance on the testing sets shows that the ANN models yielded the lowest root mean squared error for predicting RR (13.24 breaths/min), VT (0.30 °C), and ET (0.29 °C). The results interpreted with partial dependence plots and Local Interpretable Model-agnostic Explanations show that P.M. measurements and winter calving contributed most to high RR and VT predictions, whereas lying posture, high ambient temperature, and low wind speed contributed most to high ET predictions. When determining the ground-truth heat stress state by the actual RR, the best classification performance was yielded by the predicted RR with an accuracy of 77.7%; when determining the ground-truth heat stress state by the actual VT, the best classification performance was yielded by the predicted VT with an accuracy of 75.3%. This study demonstrates the ability of ANN in predicting physiological responses of dairy cows raised on actual farms with access to sprinklers. Adding more predictors other than meteorological parameters into training could increase predictive performance. Recognizing the heat stress state of individual animals, especially those at the highest risk, based on the predicted physiological responses and their interpretations can inform better heat abatement decisions. [ABSTRACT FROM AUTHOR]