A method to evaluate the behavior and well-being of pigs based on video stream analysis

Abstract

This paper describes the operation and principle of operation of the newly developed automated remote animal tracking system “Swine Track” based on the analysis of the video stream received from video surveillance cameras and the results of testing the system at the fattening site in a peasant farm enterprise. As known, pigs may experience stress at all stages of their development. For example, weaned pigs experience stress when they are weaned from their mothers, change their living conditions and type of feeding. All these stress factors may lead to fighting, injuries, diseases and deaths. Continuous monitoring of animal behavior and analysis of the data obtained will allow quickly responding to anomalies, deviations from the norm and prescribing treatment for animals, and adjusting their management conditions. According to the world experience, the use of digital technologies in agriculture can increase production capacity and economic efficiency. The automated remote animal tracking system “Swine Track” is capable of ensuring continuous monitoring. It consists of video equipment and specially developed software for detecting deviations and anomalies in animal behavior. The system identifies each animal by the numbers applied on their backs with harmless pencils, and collects information about an animal by comparing it with normative indices, and in case of deviation from the norm delivers an alerting signal. During the testing of the “Swine Track” system at the fattening site, its effectiveness was proven. Early diagnosis of changes in animal behavior contributes to their recovery and the herd survival which has a positive effect on the economic efficiency of the enterprise.