Determination of live weight of pigs from dimensions measured using image analysis

Nabil Brandl and Erik J›rgensen
Danish Institute of Agricultral Sciences,
(1) Dept. of Animal Health and Welfare,
(2) Dept. of Agricultural system,
Research Center Foulum, P.O. Box 50, DK-8830 Tjele, Denmark
E-MAIL: Nabil.Brandl@agrsci.dk

Abstract

Determination of the live weight of pigs from dimensions measured using an image analysis system has been studied in an experiment at Research Center Foulum, Denmark. The experiment included 416 crossbred pigs and compared ad libitum versus restricted feeding. Each pig was weighed and video recorded (from above) 5 times during the growing period from 25 to approx. 100 kg. The main purpose of the experiment was to estimate the precision of live weight prediction using image analysis. Spline functions (Piecewise polynomials) were used to express the relationship between body area of the pig measured by image analysis and live weight of the pig for each of the three replicates of the experiment. A mixed model was used to explain the random effect of pigs within blocks. The error expressed in term of standard deviation of the residuals from the spline function in percent of mean was estimated as 8-9%, while was estimated by mixed model as 5-6%. The relationship between body area and live weight was almost linear on a logarithmic scale, and the residual standard deviation was constant. The relationship differed between replicates presumably due to systematic error in measurement technique. Inclusion of other body dimensions only gave slight improvements in the precision. Clear differences in the size of the residuals were demonstrated for pigs from different breed combinations with Danish Yorkshire having a heavier weight than Danish Landrace at the same body area. The feeding methods had only minor influence on the relationship. Even though the precision is fairly low, image analysis techniques for weight determination of pigs seems to be promising, because it can use standard equipment with a relative low price, the possibility of using the equipment for many pigs, and finally the possibility of using the equipment for other surveillance purposes in the herd. However, further research is needed into techniques for automatic image analysis, possibly using a three-dimensional approach, and into techniques for automatic calibration of herd / breed specific area-weight relationships.

Keywords: Live weight, body dimensions, image analysis and pig production.