Year: 2024
Type: conference-abstract
Abstract: One of the constraints on advancing livestock research is the lack of large-scale quantification of physiological traits, a process often constrained by financial, labor-intensive, and scalability limitations, in addition to being traditionally performed through invasive methods. However, in the era of big data, the collection of large, extensive, and heterogeneous datasets via digital technologies, coupled with Artificial Intelligence (AI) techniques, has catalyzed significant progress across various research domains, including precision medicine, education, finance, and environmental and socioeconomic studies. This presentation will highlight our research efforts in employing computer vision systems (CVS). It will focus on a longitudinal study predicting body weight (BW) using a keypoint model. The discussion will include the advantages of using keypoints and 2D images to assess BW, which is important in the fields of animal growth and physiology in dairy and beef farm operations. We will also discuss the use of machine learning algorithms to early detect subclinical ketosis and locomotion problems in dairy cows. Lastly, this presentation will showcase the use of Unmanned Aerial Vehicles (UAVs) for pasture and soil management in grazing conditions. Our main goal with this presentation is to discuss how CVS and AI can generate valuable data in different areas, allowing for large-scale phenotyping, improving farm performance and sustainability.
Source: Journal of Animal Science
Authors: Joao Dorea, Guilherme Lobato Menezes
Institution: University of Wisconsin–Madison
Language: English
Topic: Smart Agriculture and AI
Subfield: Plant Science
Domain: Life Sciences
SDG: Zero hunger
Open Access status: green
OpenAlex ID: W4402533528