Random regression models were used to analyse the daily growth data for a total of 25 pigs of two commercial crossbred types between 75 and 140 days of age. A visual imaging system placed above a feeding station provided daily the plan area and length measurements of different body parts. Daily live-weight measurements of the pigs were obtained from a platform balance integrated into an electronic feeding station. Growth curves associated with different measures, pigs and types were compared. Significant differences in the age growth curves between the pig types could only be found in the ham width measurements (P < 0.05). The linear measure of ham width showed the greatest difference between the two types, and the lowest coefficient of variation among individual animals. Size measures were shown to be a more consistent indicator of pig performance during growth than live weight: pigs with a relatively large surface area or ham width at the early growth stage also have relatively large surface area or ham width at later stages and the between-animal variation in these measurements remains constant with age. Gain in live weight relative to increase in size differed significantly between the two pig types (P < 0.05). Pigs of the two types had significantly different shapes, but the change of shape during growth did not differ significantly between them. The allometric relationships between surface area and ham width1.85 and between body length and ham width0.85 indicate that the ham widths of pigs increase faster in proportion to full body measures. Variations between individual animals in size increase and shape change are significant (P < 0.05). The analysis suggests that VIA size and shape measurements provide valid descriptors of pig growth.

A visual image analysis (VIA) system provided continuous, automatic collection of size and shape data for a total of 116 pigs slaughtered serially from 25 to 115 kg live weight. Males and females of three types of pigs (‘Meishan’ type, ‘Pietrain’ type, and ‘Landrace’ type) were selected to provide variation in both composition and conformation (the three types being, respectively, ‘fat’, ‘blocky’, and ‘lean’). Results below are presented in this order. Regression analysis was used to relate VIA size to platform weigher (FIRE) measurements of live weight. Residual maximum likelihood (REML) analysis showed that at the observed growth rate, a change in pig state could be detected by VIA after 8, 9, and 10 days respectively for the three types, and by the platform weigher system after 12, 4, and 13 days (in both cases with a confidence of 95%). Artificial neural network and canonical variates analysis were used to test the ability of VIA to distinguish between pig types and sexes. With cross validation, the canonical variates analysis correctly classified the three types in 72, 83, and 64% of observations, and the neural network in 81, 81, and 64% of observations. The VIA system is considered to be a valuable monitoring system which may play a rôle in the construction of integrated management systems (IMS).

The conformation, tissue composition, and chemical composition of three types of pigs, given food ad libitum and slaughtered over a nominal live weight range of 35 to 115 kg, was assessed in relation to data provided on the live animals by a visual image analysis (VIA) system. The pig types were named as ‘3⁄4 Landrace’, ‘1⁄2 Pietrain’, and ‘1⁄4 Meishan’ types, representing ‘attenuated’, ‘blocky’, and ‘flabby’ types. Three analyses of the shape, conformation and composition data were performed. First, the relationship between conformation and age/size was assessed using linear regression of logarithmically transformed VIA and carcass data. In relation to age, ‘1⁄2 Pietrain’ pigs were found by both VIA and carcass measurements to have the widest shoulders. Both analyses also found this type to have the widest ham, trunk, and shoulders in relation to body length across most of the body length range studied, although the greatest rate of increase in ham width in relation to body length was found in the ‘1/4 Meishan’ type pigs. Second, the relationship between composition and VIA shape was examined using linear regression of transformed and standardized data. Significant relationships were found between fat, lipid, muscle, and protein weight and VIA shape, although relationships were weaker for protein and muscle weight. For fat and lipid, the VIA shape measures from the trunk region proved the most informative, whereas the VIA ham measures proved the most informative for muscle and protein. Third, detrended measures of composition/conformation and shape were used to remove the effect of animal size from the data. Removal of the variation due to growth generally led to substantial decreases in the adjusted R2 statistics and in the R2-like statistics for prediction. Although in the models without detrending, relative fat and lipid weight had been found most strongly correlated with VIA shape, relative muscle was found most strongly correlated with shape in the detrended data. This was considered to result from the low between-animal variation in the data set combined with greater across-weight variation in fat and lipid weights than muscle and protein weights in the data without trend removal. Future trials with greater between-animal variation imposed would allow more precise determination of the relationship between conformation and shape.