Citation format: Zhang Kai, Han Shuqing, Cheng Guodong, Wu Saisai, Liu Jifang. Identification of Cow Gait Phase Based on Gaussian Hybrid-Hidden Markov Fusion Algorithm[J]. Smart Agriculture, 2022, 4(2): 53-63.
ZHANG Kai, HAN Shuqing, CHENG Guodong, WU Saisai, LIU Jifang. Gait phase recognition of dairy cows based on Gaussian Mixture model and Hidden Markov model[J]. Smart Agriculture, 2022, 4(2): 53-63.
Click DIRECT TO CNKI to read
Click directly to the official website to read (full text, free)
Based on Gaussian hybrid-hidden Markov fusion algorithm, the cow gait phase is identified
KAI ZHANG,SHUqing HAN,GUOdong CHENG,SAISAI WU,JIfang LIU*
(Institute of Agricultural Information, Chinese Academy of Agricultural Sciences/Key Laboratory of Blockchain Agriculture Application, Ministry of Agriculture and Rural Affairs, Beijing 100081, China)
Abstract: Cow gait phase is an important indicator of cow health and lameness. In order to accurately and automatically identify the cow gait phase, this study proposes an unsupervised learning cow gait phase recognition algorithm GMM-HMM that integrates Gaussian Mixture Model (GMM) and Hidden Markov Model (HMM). Using the inertial measurement unit to collect the acceleration and angular velocity signals of the cow's hind limbs, the Kalman filter eliminates noise, screens and extracts the eigenvalues, and constructs the GMM-HMM model to automatically identify the three gait phases of the cow, such as the static phase, the standing phase and the oscillating phase in the continuous gait. The results showed that the accuracy rate, recall rate and F1 of The Static Acquaintance were 89.28%, 90.95% and 90.91%, respectively, the accuracy rate, recall rate and F1 of the Standing Acquaintance in the continuous gait were 91.55%, 86.71% and 89.06%, respectively, and the accuracy rate, recall rate and F1 of the Swing Acquaintance in the continuous gait were 86.67%, 91.51% and 89.03%, respectively. The accuracy of cow gait segmentation was 91.67%, which was 4.23% and 1.1% higher than that of event-based peak detection method and dynamic time regularization algorithm, respectively. This study can provide a technical reference for the next step of dairy cow lame feature extraction based on wearable gait analysis.
Keywords: cow lame; Gait phase; Gait segmentation; Gaussian hybrid model; Hidden Markov model; Kalman filtering
Image of the article
Figure 1 Schematic diagram of the measurement site
Fig. 1 Schematic diagram of measuring site
Figure 2 Sensor wearing method and sensor coordinates
Fig. 2 Sensor wearing method and sensor coordinates
Note: AX is the x-axis acceleration and GZ is the z-axis angular velocity
Figure 3 The cow gait events corresponding to the x-axis acceleration and z-axis angular velocity line charts
Fig. 3 Gait events of dairy cows corresponding to the curve of x-axis acceleration and z-axis angular velocity
Note: Acceleration of the X, AY and AZ axes of x, y and z, respectively; The angular velocities of the x, GY, and GZ axes are x, y, and z, respectively
Fig. 4 Sensor 6-axis data line chart
Fig. 4 Sensor 6-axis data line chart
Note: Si is the hidden state, Oi is the observation state, i=1,2,3... n
Figure 5 Hidden Markov model structure
Fig. 5 Hidden Markov Model structure
Figure 6 The process of state transition of the hidden Markov model
Fig. 6 State transition processof Hidden Markov Model
Figure 7 Flowchart of adaptive identification method for the gait phase of cows
Fig. 7 Flow chart of adaptive recognition method for gait phase of dairy cows
Figure 8 Mathematical characteristics of different phases of cow gait
Fig. 8 Mathematical characteristics of different gait phases of dairy cows
Fig. 9 Results of phase recognition of cow gait
Fig. 9 Gait phase recognition results of dairy cows
Source: Smart Agriculture (Chinese and English), No. 2, 2022
Please contact the editorial board for authorization
About the Author of the Newsletter
Liu Jifang, Researcher
Liu Jifang is the Secretary of the Party Committee and Deputy Director of the Institute of Information of the Chinese Academy of Agricultural Sciences, a researcher and doctoral supervisor. Chief Scientist of the Internet of Things Team of Agricultural Network of The Chinese Academy of Agricultural Sciences. He is a member of the Agricultural Informatization Standardization Technical Committee of the Ministry of Agriculture and Rural Affairs, a vice chairman of the Computer Application Branch of the Chinese Agricultural Society, and the director of the editorial board of Smart Agriculture (Chinese and English). He has long been engaged in the research of agricultural informatization and sustainable development of agriculture. He has presided over and participated in more than 40 national science and technology key projects and the National Natural Science Foundation of China. He has published more than 60 research papers and 6 monographs. It has won 6 awards for scientific and technological achievements at the national, provincial and ministerial levels.
Supporting units for this issue
Jinglan Yunzhi Internet of Things Technology Co., Ltd
Zhejiang Zhenshan Technology Co., Ltd
Weichai Revo Heavy Industry Co., Ltd
small
shop
Welcome to shop
WeChat communication service group
In order to facilitate the academic exchange of readers, authors and review experts in the field of agricultural science, promote the development of smart agriculture, and better serve the majority of readers, authors and reviewers, the editorial department has established a WeChat exchange service group, and the discussion of issues in the professional field and the issues related to submission can be consulted in the group.
How to join the group: add Xiaobian WeChat 331760296, Remarks: name, unit, research direction, Xiaobian pull you into the group, institutional marketing advertisers do not disturb.
Publication of Information
Introduction of scientific research team and promotion of recruitment information, academic conferences and related activities