In this paper, the dairy cattle movement detecting technology based on 3-axis acceleration sensor information fusion is presented. For they show ideal performance in generalization and optimization, Support vector machines are used to build an information fusion model for dairy cattle’s behavior classification. The data feature of the support vector machine fusion model is derived from 3-axis acceleration data. RBF function is used as the model’s kernel function. The genetic algorithm is used to optimize the parameters of the kernel function. The training and testing results show that using genetic algorithm for kernel function parameter searching has good ability to optimize the fusion model.
Keywords
Dairy cattle movement detection, support vector machine, genetic algorithm
Thank you. But , these papers are talking about how to detect the dairy cow motion detection. I need a link to download a standard dataset about that for my research.
I thought you wanted to conduct experiments on cow behavior and need to know standard techniques. The behavior of each individual cow may be different and if you study a large (statistically designed/determined) number of cows belonging to a particular breed/age group/ location/ feeding pattern etc. than you can generalize your observed data specifically for your studied animal group (or those belonging to that specific category of animals elsewhere under similar conditions) and based on your study, you can rate those animals normal or abnormal as per the statistical parameters that you ascertain through your observations.
You are right. The behavior of human beings are different too, but we have several standard datasets for human motion detection using three-axis acceleration. Is there any database for animals motion detection with a download link?