AI tool estimates cattle temperature from a thermal photo

Researchers at the University of Arkansas have developed CattleFever, an AI system that uses thermal images of a calf’s face to estimate body temperature without rectal thermometry. In testing, the tool came within about 1 degree of thermometer readings, using heat patterns around the eyes and nostrils as the strongest predictors. The work was published in Smart Agricultural Technology, and the team also released an open paired RGB-thermal dataset, CattleFace-RGBT, to support follow-on research. (arkansasresearch.uark.edu)

Why it matters: For veterinary professionals and herd health teams, the promise is lower-stress temperature screening that could reduce handling time, labor, and risk to staff while making it easier to monitor more animals more often. That said, this is still an early-stage research tool, not a field-ready replacement for rectal thermometry: the current system was developed from images taken with calves facing the camera in controlled conditions, and prior cattle thermography research has shown that infrared readings can be useful for screening but don’t always cleanly substitute for core temperature in real-world use. (arkansasresearch.uark.edu)

What to watch: The next step is validation in less controlled farm settings, including varied angles, movement, and larger herds, before the technology is likely to move into practical on-farm monitoring. (arkansasresearch.uark.edu)