New Swedish research shows that artificial intelligence and smartphone technology can be used to estimate the weight of dairy cows without the need for scales or measuring tapes. Researchers believe the technology could support health monitoring, feed management and animal welfare.
According to a press release, researchers at the Swedish University of Agricultural Sciences (SLU) have developed an AI-powered app capable of estimating the weight of dairy cows using a smartphone.
The app combines artificial intelligence, image analysis and LiDAR technology to calculate a cow’s weight from images captured on a mobile phone. According to the researchers, the technology could make it easier to collect important information about animal health and development without the need for weighing scales or measuring tapes.
The project was led by Oleksiy Guzhva, Associate Professor at the Department of Biosystems and Technology at SLU.
“It sounded complicated, but we couldn’t verify the claims. Having worked with image analysis for many years, I wanted to investigate whether the technology could work. If successful, it could make life easier for farmers,” he said.
To develop the system, the researchers collected more than 10,000 images of dairy cows from different angles and distances in commercial barns. Using AI, the system identifies up to 600,000 reference points on each animal, enabling accurate body measurements and weight estimates.
Applications in herd health and feed management
According to the researchers, precise weight estimates could support a range of on-farm decisions, including feed management, health monitoring and veterinary treatment, where accurate weight information is essential for determining medicine dosages.
“With this technology, farmers no longer need to use measuring tapes and scales, which can be time-consuming and may increase the risk of injury. We want the app to simplify difficult tasks and provide valuable information,” said Oleksiy Guzhva.
The research team also sees potential for the technology to support the collection of health and welfare indicators across dairy herds. In the longer term, the method could be extended to other cattle, calves and horses.
The findings have been published in Nature under the title “PickAMoo: LIDAR-enhanced mask R-CNN segmentation for precision weight estimation in dairy cattle using smartphone imaging.”
By Lene Mikkelsen Walsh – edited by Maja Løvstrup
Photo: SLU / Oleksiy Guzhva
