Skip to main content
ARS Home » Plains Area » Lubbock, Texas » Cropping Systems Research Laboratory » Livestock Issues Research » Research » Publications at this Location » Publication #193706


item Dailey, Jeffery - Jeff
item Krebs, Nadege
item Carroll, Jeffery - Jeff Carroll
item Mcglone, John

Submitted to: American Society of Animal Science Annual Meeting
Publication Type: Abstract Only
Publication Acceptance Date: 2/23/2006
Publication Date: 7/9/2006
Citation: Dailey, J.W., Krebs, N., Carroll, J.A., McGlone, J., 2006. Validation of a color automated tracking system for activity and pen location of group housed weanling pigs [abstract]. Journal of Animal Science. 84:413(Suppl. 1). Abstract #519.

Interpretive Summary:

Technical Abstract: Knowing where animals are located throughout the day can be important in some studies. While scan sampling methods can be imprecise, continuous video observation is precise but time-consuming. Commercial automated tracking systems (ATS) quantify animal behavior by assigning an X, Y coordinate for an animal's location at a given time. The objective of this study was to validate an ATS, Ethovision, using the 'color' tracking method compared with a human observer using the Observer 5.0 (HOB). Nursery pigs (n = 44, 1 pig was removed from the study due to technical problems with the HOB) were group housed (n = 4/pen). Each pig in the pen had a different color tape around its shoulders (11 replications/color). Pigs were video recorded continuously for 1 h with color cameras at 30 frames/s. Videos were scored for a 1 h period for the time spent on the right half of the pen. Data obtained with the ATS were compared to data collected by HOB by ANOVA and regression analyses. The following results were obtained using 10 pixels as a parameter of detection in ATS. No difference (P=0.597) was observed between HOB and ATS for the time spent on the right. A linear regression comparing the methods of observation for 'time spent on the right' yielded an r2 of 0.9102 (or r = 0.954, P < 0.01) which indicated ATS as an assay was a precise predictor of HOB measurements (using 100 pixels, there was no difference in the means but the r2 was 0.78). The simple linear regression model was HOB = 0.962 ATS + 3.6929 (SEb = 0.047). The number of pixels used in the ATS must be optimized to accurately interpret animal movement. Additionally, the surface area covered by the tape, the color and elasticity of the tape, and the lighting system in the room can also affect the efficiency of the ATS. In conclusion, the ATS generated mean values similar to behavioral data collected by HOB. The r2 demonstrates that the ATS closely predicted data obtained by HOB. The ATS is recommended for collection of this type of behavioral information.