19 How can Artificial Intelligence Accelerate Phenotyping Efforts in Animal Breeding?. (21st September 2022)

• Record Type: Journal Article
• Authors: Morota, Gota; Ha, Dong; Chen, James
• Is Part Of: Journal of animal science. Volume 100(2022)Supplement 3
• Page Start: 11
• Page End: 12
• Publication Date: 2022-09-21
• Subjects: computer vision -- phenotyping -- wearable sensor; Livestock -- Periodicals; Livestock; Electronic journals; Periodicals; 636.005
• DOI: 10.1093/jas/skac247.020
• Languages: English
• ISSNs: 0021-8812

Abstract

Abstract: With the development of high-throughput technologies, biology has become a large-scale and data-rich field, especially for genomics and phenomics. Concurrent with this expansion in the availability of high-throughput technologies, their use in the livestock sectors has accordingly increased. This expansion has occurred most notably in the field of animal breeding, where high-throughput technologies hold promise as a means for more efficiently connecting animal phenotypes with pedigree and genomics to drive genetic improvement of production and health-related traits. The rate of genetic gain is closely related to the quantity and quality of phenotyping data. However, there are still considerable manual tasks that are involved in phenotyping processes. Precision livestock farming or smart farming uses sensing technology that is supported by artificial intelligence and machine learning to monitor morphometric changes in animal growth dynamics and animal activity status. In particular, the development of two sensing technologies, computer vision and wearable sensor systems, plays a pivotal role in accelerating phenotyping efforts by providing non-intrusive measurements of animals with high temporal and spatial resolution. The presentation will provide a recent update on current approaches to artificial intelligence and machine learning in computer vision and wearable sensor systems to monitor the body mass and behaviors of animals.