Detecting Dairy Cow Behavior Using Vision Technology

The aim of this study was to investigate using existing image recognition techniques predict the behavior dairy cows. A total 46 individual cows were monitored continuously under 24 h video surveillance prior calving. annotated for behaviors standing, lying, walking, shuffling, eating, drinking and contractions each cow from 10 19,191 records obtained a non-local neural network trained validated on clips behavior. This showed that used correctly classified seven 80% or more time in dataset. In particular, detection birth predicted 83% time, which itself can be an early warning calving alert, as all start several hours giving birth. approach cameras assist livestock management.