PSIII-7 Analysis of sow postures using infrared sensors

Abstract

Abstract The health of the sow is crucial in determining the economic success of swine producers. Sows spend most of their time standing, and due to their heavy weight, they are exposed to more leg injuries compared to other livestock. Frequent observations are required to examine the body structure of sows to provide appropriate treatment, but such observations are done manually by technicians. Considering the lack and high price of labor, an automated investigation system that examines the body structure would greatly benefit sow farms. Various technologies for animal health have been introduced and used in the livestock field to construct a successful smart farm. This study aims to present an assessment method applied to an individual sow using sensor technology, suggesting a system to observe various postures of sows, which help producers decide leg injuries or bad health symptoms in real time. The system consists of 8 infrared distance sensors: three that point from the front, two from the middle, and three from the rear side. To compare the postures of normal and abnormal sows, one sow was normally treated and the other was injected with 70% of dextrose at 9 am every day. During 28 days of experimentation, the normal sow showed 39 minutes of movement on average per day for 8 days at sensor 3 and 97 minutes a day for 17 days at sensor 5. The abnormal sow showed 97 minutes a day for 21 days at sensor 3 and 215 minutes a day for 27 days at sensor 5, which were significantly different from the normal sow (P < 0.05). Simultaneous movements at low levels of the front and middle sensors could help differentiate between normal and abnormal individuals. Such a promising system could serve as a vital indicator for animal health and maximize livestock farm revenue.