Cow behavior monitoring is critical for understanding the current state of cow welfare and developing an effective planning strategy for pasture management, such as early detection of disease and estrus. One of the most powerful and cost-effective methods is a neural-network-based monitoring system that analyzes time series data from inertial sensors attached to cows. For this method, a significant challenge is to improve the quality and quantity of teaching data in the development of neural network models, which requires us to collect data that can cover various realistic conditions and assign labels to them. As a result, the cost of data collection is significantly high. This work proposes a data augmentation method to solve two major quality problems in the collection process of teaching data. One is the difficulty and randomicity of teaching data acquisition and the other is the sensor position changes during actual operation. The proposed method can computationally emulate different rotating states of the collar-type sensor device from the measured acceleration data. Furthermore, it generates data for actions that occur less frequently. The verification results showed significantly higher estimation performance with an average accuracy of over 98% for five main behaviors (feeding, walking, drinking, rumination, and resting) based on learning with long short-term memory (LSTM) network. Compared with the estimation performance without data augmentation, which was insufficient with a minimum of 60.48%, the recognition rate was improved by 2.52-37.05pt for various behaviors. In addition, comparison of different rotation intervals was investigated and a 30-degree increment was selected based on the accuracy performances analysis. In conclusion, the proposed data expansion method can improve the accuracy in cow behavior estimation by a neural network model. Moreover, it contributes to a significant reduction of the teaching data collection cost for machine learning and opens many opportunities for new research.
Chao LI
Tokyo Institute of Technology
Korkut Kaan TOKGOZ
Tokyo Institute of Technology
Ayuka OKUMURA
Shinshu University
Jim BARTELS
Tokyo Institute of Technology
Kazuhiro TODA
Information Services International-Dentsu, Ltd.
Hiroaki MATSUSHIMA
Information Services International-Dentsu, Ltd.
Takumi OHASHI
Tokyo Institute of Technology
Ken-ichi TAKEDA
Shinshu University
Hiroyuki ITO
Tokyo Institute of Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Chao LI, Korkut Kaan TOKGOZ, Ayuka OKUMURA, Jim BARTELS, Kazuhiro TODA, Hiroaki MATSUSHIMA, Takumi OHASHI, Ken-ichi TAKEDA, Hiroyuki ITO, "A Data Augmentation Method for Cow Behavior Estimation Systems Using 3-Axis Acceleration Data and Neural Network Technology" in IEICE TRANSACTIONS on Fundamentals,
vol. E105-A, no. 4, pp. 655-663, April 2022, doi: 10.1587/transfun.2021SMP0003.
Abstract: Cow behavior monitoring is critical for understanding the current state of cow welfare and developing an effective planning strategy for pasture management, such as early detection of disease and estrus. One of the most powerful and cost-effective methods is a neural-network-based monitoring system that analyzes time series data from inertial sensors attached to cows. For this method, a significant challenge is to improve the quality and quantity of teaching data in the development of neural network models, which requires us to collect data that can cover various realistic conditions and assign labels to them. As a result, the cost of data collection is significantly high. This work proposes a data augmentation method to solve two major quality problems in the collection process of teaching data. One is the difficulty and randomicity of teaching data acquisition and the other is the sensor position changes during actual operation. The proposed method can computationally emulate different rotating states of the collar-type sensor device from the measured acceleration data. Furthermore, it generates data for actions that occur less frequently. The verification results showed significantly higher estimation performance with an average accuracy of over 98% for five main behaviors (feeding, walking, drinking, rumination, and resting) based on learning with long short-term memory (LSTM) network. Compared with the estimation performance without data augmentation, which was insufficient with a minimum of 60.48%, the recognition rate was improved by 2.52-37.05pt for various behaviors. In addition, comparison of different rotation intervals was investigated and a 30-degree increment was selected based on the accuracy performances analysis. In conclusion, the proposed data expansion method can improve the accuracy in cow behavior estimation by a neural network model. Moreover, it contributes to a significant reduction of the teaching data collection cost for machine learning and opens many opportunities for new research.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.2021SMP0003/_p
Copy
@ARTICLE{e105-a_4_655,
author={Chao LI, Korkut Kaan TOKGOZ, Ayuka OKUMURA, Jim BARTELS, Kazuhiro TODA, Hiroaki MATSUSHIMA, Takumi OHASHI, Ken-ichi TAKEDA, Hiroyuki ITO, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={A Data Augmentation Method for Cow Behavior Estimation Systems Using 3-Axis Acceleration Data and Neural Network Technology},
year={2022},
volume={E105-A},
number={4},
pages={655-663},
abstract={Cow behavior monitoring is critical for understanding the current state of cow welfare and developing an effective planning strategy for pasture management, such as early detection of disease and estrus. One of the most powerful and cost-effective methods is a neural-network-based monitoring system that analyzes time series data from inertial sensors attached to cows. For this method, a significant challenge is to improve the quality and quantity of teaching data in the development of neural network models, which requires us to collect data that can cover various realistic conditions and assign labels to them. As a result, the cost of data collection is significantly high. This work proposes a data augmentation method to solve two major quality problems in the collection process of teaching data. One is the difficulty and randomicity of teaching data acquisition and the other is the sensor position changes during actual operation. The proposed method can computationally emulate different rotating states of the collar-type sensor device from the measured acceleration data. Furthermore, it generates data for actions that occur less frequently. The verification results showed significantly higher estimation performance with an average accuracy of over 98% for five main behaviors (feeding, walking, drinking, rumination, and resting) based on learning with long short-term memory (LSTM) network. Compared with the estimation performance without data augmentation, which was insufficient with a minimum of 60.48%, the recognition rate was improved by 2.52-37.05pt for various behaviors. In addition, comparison of different rotation intervals was investigated and a 30-degree increment was selected based on the accuracy performances analysis. In conclusion, the proposed data expansion method can improve the accuracy in cow behavior estimation by a neural network model. Moreover, it contributes to a significant reduction of the teaching data collection cost for machine learning and opens many opportunities for new research.},
keywords={},
doi={10.1587/transfun.2021SMP0003},
ISSN={1745-1337},
month={April},}
Copy
TY - JOUR
TI - A Data Augmentation Method for Cow Behavior Estimation Systems Using 3-Axis Acceleration Data and Neural Network Technology
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 655
EP - 663
AU - Chao LI
AU - Korkut Kaan TOKGOZ
AU - Ayuka OKUMURA
AU - Jim BARTELS
AU - Kazuhiro TODA
AU - Hiroaki MATSUSHIMA
AU - Takumi OHASHI
AU - Ken-ichi TAKEDA
AU - Hiroyuki ITO
PY - 2022
DO - 10.1587/transfun.2021SMP0003
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E105-A
IS - 4
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - April 2022
AB - Cow behavior monitoring is critical for understanding the current state of cow welfare and developing an effective planning strategy for pasture management, such as early detection of disease and estrus. One of the most powerful and cost-effective methods is a neural-network-based monitoring system that analyzes time series data from inertial sensors attached to cows. For this method, a significant challenge is to improve the quality and quantity of teaching data in the development of neural network models, which requires us to collect data that can cover various realistic conditions and assign labels to them. As a result, the cost of data collection is significantly high. This work proposes a data augmentation method to solve two major quality problems in the collection process of teaching data. One is the difficulty and randomicity of teaching data acquisition and the other is the sensor position changes during actual operation. The proposed method can computationally emulate different rotating states of the collar-type sensor device from the measured acceleration data. Furthermore, it generates data for actions that occur less frequently. The verification results showed significantly higher estimation performance with an average accuracy of over 98% for five main behaviors (feeding, walking, drinking, rumination, and resting) based on learning with long short-term memory (LSTM) network. Compared with the estimation performance without data augmentation, which was insufficient with a minimum of 60.48%, the recognition rate was improved by 2.52-37.05pt for various behaviors. In addition, comparison of different rotation intervals was investigated and a 30-degree increment was selected based on the accuracy performances analysis. In conclusion, the proposed data expansion method can improve the accuracy in cow behavior estimation by a neural network model. Moreover, it contributes to a significant reduction of the teaching data collection cost for machine learning and opens many opportunities for new research.
ER -