Multi-source information fusion dairy cow behavior monitoring system and method

Abstract

The invention provides a multi-source information fusion dairy cow behavior monitoring system and method. The method comprises the steps of performing data fusion on collected body temperature parameters, pulse parameters and exercise amount parameters to obtain body temperature data, pulse data and exercise amount data; collecting video frames in dairy cow monitoring videos, and performing normalization conversion on the collected video frames to obtain image data; on the basis of the body temperature data, the pulse data, the exercise amount data and the image data, constructing a sample data matrix; on the basis of the sample matrix, adopting a density clustering algorithm for determining whether an outlier exists or not; if the outlier exists, indicating that dairy cow is in the heat state at an outlier corresponding moment. By means of complementarity among multi-dimensional data, the problem that single data is poor in anti-jamming capability is avoided, accurate judgment of thedairy cow behavior state is achieved, and the effects of high reliability and good fault tolerance are achieved.

Description

technical field [0001] The invention relates to the technical field of monitoring, in particular to a multi-source information fusion dairy cow behavior monitoring system and method. Background technique [0002] With the progress and development of the times, people's living standards are gradually improving, and the demand for dairy products in daily life is also increasing. The key factor for the development of the dairy industry is the output and quality of raw milk, and the health status and estrus of dairy cows are the key factors that determine the quality and output of milk. The physical parameters of dairy cows include body temperature, pulse, exercise volume, etc., which are important indicators for evaluating the health and physiological state of dairy cows. Under normal circumstances, the physical parameters of dairy cows only change slightly within a relatively constant range, but they will increase or decrease to varying degrees during pathological processes a...

AI Technical Summary

Problems solved by technology

However, the current digital monitoring method is a single signal preprocessing technology, which only has a good information response to a certain signal, which will cause certain judgment errors, poor robustness, and a high false alarm rate and missing information. rate of return

Images