Classification of the cattle´s behaviors by using accelerometer data with simple behavioral technique

To prognosticate the cattle´s health, the farmer can observe the cattle activities such as the time period of walking-grazing, standing and sleeping. However, to monitor the cattle´s behaviors, it is unable to monitor such behavior all the time and thorough, especially raise many cattle. Therefore, this paper proposes to classify the cattle´s behaviors by using the magnitude and standard deviation of accelerometer output signal. The magnitude of each axis is used to classify the behaviors into two groups: 1) walking-grazing and standing and 2) lying down. While the standard deviation of Y-axis is used to notify the behaviors of walking-grazing and standing. The classification results were tested with two cattle and measured precise time of each behavior comparing with human observers. As a result, duration of each behavior is nearby, it has the errors as follows walking-grazing maximum errors 2% standing maximum errors 13% and lying maximum errors 7%.