Computer vision helps monitor welfare on poultry farms
The United States Department of Agriculture (USDA) is going to give a grant of one million dollars to a project proposed by the Institute of Agriculture of the University of Tennessee to create and implement a computer vision system that serves to monitor the poultry production. As the researchers explain, precision farming systems have been implemented in larger animals for some time to monitor aspects such as activity, movement and other indicators of stress and animal welfare. However, it is more difficult to monitor poultry, as they are much smaller and their populations are higher in production plants.
This project involves the creation of a computer vision system to track Animal Based Measurements (ABM) for poultry in real time. Using deep learning algorithms to identify individual birds, the program will track comfort behaviors related to welfare, such as stretching, preening, and dust bathing, as well as others related to production, such as eating and drinking. With them, the researchers will later develop a reference database with detailed notes on the behavior of the birds.
The project is estimated to cost around $2,500 per house, but the machine vision system will be practical and affordable for broiler farmers. Artificial intelligence and video image analysis will allow researchers to examine the animals' interactions with management factors and collect baseline data.
An affordable system
Lead researcher Yang Zhao, an assistant professor in the Department of Animal Science at the University of Tennessee Institute of Agriculture, explained that "despite the great interest in precision farming for poultry farming around the world, Few systems have been developed for commercial production environments. This project supports us in developing an affordable system that can help broiler producers automatically collect behavioral responses from birds and better manage flocks on commercial farms," and trusts that the collaboration of researchers, production specialists, and industry support can help develop it. The work of Zhao and his team began this February, and they hope to show their results in 2025.
Comments
Copyright © 2026 - All Rights Reserved
ISSN 2768-198X
List
Add
Please enter a comment