Skip to main content
ARS Home » Plains Area » Clay Center, Nebraska » U.S. Meat Animal Research Center » Genetics and Animal Breeding » Research » Publications at this Location » Publication #380374

Research Project: Identifying Genomic Solutions to Improve Efficiency of Swine Production

Location: Genetics and Animal Breeding

Title: Automatic phenotyping of activity traits utilizing NUtrack to enhance gilt selection

item OSTRAND, LEXI - University Of Nebraska
item TRENHAILE-GRANNEMANN, MELANIE - University Of Nebraska
item PSOTA, ERIC - University Of Nebraska
item Rohrer, Gary
item SCHMIDT, TYE - University Of Nebraska
item MOTE, BENNY - University Of Nebraska

Submitted to: European Conference on Precision Agriculture Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 1/31/2022
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: A major decision of a sow operation lies in the identification of which gilts to retain given the importance of sow longevity. Automatic computer vision allows producers to classify animals without human interference of natural behaviors. This investigation studied overall activity of replacement gilts and the use of these activities to aid in gilt retention. Beginning around 20 weeks of age, video on gilts (n = 2,374) was collected for nine consecutive d and processed using the NUtrack System, which tracks distance travelled (m), average speed (m/s), angle rotated (degrees), and time standing (s), sitting (s), eating (s), and laying (s). NUtrack is a deep learning-based multi-object tracking system that has been shown to achieve >92.5% precision and recall when tracking the long-term location and identity of individual pigs in group-housed settings. Gilts (n = 1,049) were culled based on structural unsoundness as determined by an experienced herdsman. Data were analyzed using logistic regression (RStudio V1.2.5033) with farrowing group, pen, and on-test date included in the model. Angle (P < 0.01), avg speed (P < 0.001) and standing (P < 0.001) were significantly associated with gilt retention. Heritabilities were estimated in ASReml 4.1 using an animal model with a two-generation pedigree. Heritabilities are 0.32 ± 0.048, 0.32 ± 0.049, 0.23 ± 0.044, 0.34 ± 0.051, 0.26 ± 0.046, 0.31 ± 0.049, and 0.21 ± 0.044 for average speed, distance, stand, sit, eat, angle, and laying respectively. These data suggest that animal activity and movement, as measured by NUtrack, can enhance herdsman efforts in making culling decisions of breeding animals.