Rumination recognition method of dairy cows based on the change of noseband pressure

Rumination is closely related to ruminant's health and welfare, so it is very important to monitor rumination. Rumination monitoring equipment based on chewing and swallowing audio is easy to be affected by ambient noise, equipment based on acceleration sensor is not easy to monitor number of cuds and other information, equipment based on video monitoring has high cost and is easy to be affected by weather and other cows, while equipment based on chewing pressure can accurately record the chewing pressure of dairy cows, and use related processing technology to obtain the number of cuds and other information. In the related research, the peak value and interval of masticatory pressure peak in time domain were used to identify rumination. In view of the problem that the standard of pressure peak was not well defined due to the different head size and masticatory strength of different age and different breed cows, this paper proposed a method to identify the number of rumination, duration of rumination and number of cuds of cows by using standard deviation and noseband pressure signal frequency spectrum. The accurate and stable frequency spectrum range of rumination noseband pressure signal was obtained by frequency spectrum analysis, and the interference of feeding noseband pressure signal to rumination noseband pressure signal was eliminated by using standard deviation. Compared with the direct observation, the results showed that the recognition accuracy of the method proposed in this paper were 100%, 94.2% and 94.45% for the number of rumination, duration of rumination and number of cuds respectively, and the average absolute errors were 0, 2.0029% and 2.5623% respectively, which showed that the method can accurately identify the rumination information of dairy cows.