
Research Article
Study on Ruminant Recognition of Cows Based on Activity Data and Long Short-Term Memory Network
@INPROCEEDINGS{10.1007/978-3-030-62205-3_13, author={Shuai Hou and Xiaodong Cheng and Mingshu Han}, title={Study on Ruminant Recognition of Cows Based on Activity Data and Long Short-Term Memory Network}, proceedings={Mobile Wireless Middleware, Operating Systems and Applications. 9th EAI International Conference, MOBILWARE 2020, Hohhot, China, July 11, 2020, Proceedings}, proceedings_a={MOBILWARE}, year={2020}, month={11}, keywords={Rumination of cows Activity data Long Short-Term memory}, doi={10.1007/978-3-030-62205-3_13} }
- Shuai Hou
Xiaodong Cheng
Mingshu Han
Year: 2020
Study on Ruminant Recognition of Cows Based on Activity Data and Long Short-Term Memory Network
MOBILWARE
Springer
DOI: 10.1007/978-3-030-62205-3_13
Abstract
In this paper, the collected activity data with ruminating status label is used as the data set, based on the long short-term memory network in the recurrent neural network, in order to identify and judge the ruminating process of dairy cows. This paper analyzes the advantages of selecting activity data as input data and long short-term memory network as core algorithm, introduces the hardware design and composition of the self-developed activity data acquisition system, and describes the characteristics of long short-term memory network structure. It is innovative to combine cow activity data with long short-term memory network to identify ruminating in the time period of cow activity data. The experimental results show that the long short-term memory network has different recognition effects on dairy cows of different individuals through the learning of activity data, and the accuracy of ruminating recognition of the whole data is 0.78. This method is effective and feasible. It can provide ideas for the related research of intelligent animal husbandry.