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%.