E. Vranken

1. The objective was to investigate the possibility of detecting daily body weight changes of broiler chickens with computer-assisted image analysis. 2. The experiment included 50 broiler chickens reared under commercial conditions. Ten out of 50 chickens were randomly selected and video recorded (upper view) 18 times during the 42-d growing period. The number of surface and periphery pixels from the images was used to derive a relationship between body dimension and live weight. 3. The relative error in weight estimation, expressed in terms of the standard deviation of the residuals from image surface data was 10%, while it was found to be 15% for the image periphery data. 4. Image-processing systems could be developed to assist the farmer in making important management and marketing decisions.

Growth control in broiler chickens has been proven to be an efficient method to ensure broiler performance and yield and to lessen health problems. The growth control procedure has been tested in literature using a limited amount of animals in experimental facilities. Under these "ideal" circumstances, the birds could follow predefined growth trajectories with accuracies ranging from 3.7 to 6% (mean relative error). The objective of this research was to test the above growth control procedure in the field in a real broiler farm, evaluate its accuracy, and explore its benefits for the broiler farmer. In this procedure, a model-based control algorithm was used to calculate the feed supply to the broilers with the intention of following a target growth trajectory as close as possible. A simultaneous small-scale experiment was performed to have an idea about the order of magnitude of the accuracy of the same procedure under "ideal" laboratory conditions. In farm conditions, the mean relative error between the target trajectory and the weight of the controlled birds was 7.3%. Higher than that under laboratory conditions (2.7%), it was indicative of the challenges of transferring the growth control procedure to real scale. On the other hand, the growth control procedure has been proved to be beneficial under the farm conditions in the feed conversion ratio. The feed conversion ratio of the controlled group was better (1.64) than the ad libitum-fed ones (1.68).

Applying altered trajectories in broiler growth control with early feed restriction and a consequent accelerated catch-up growth has been approved to result in a better feed conversion ratio and a reduction in mortality. The properties of the growth trajectory and the resulting time and duration of the feed restriction can be crucial for animal welfare and production performance. The objective of this work was to test broiler growth control strategy online in field conditions using different target trajectories. Several experiments were conducted, and the best target trajectory has been proven to result in an end weight of 2,616 g and feed conversion ratio of 1.54 for Ross-type birds and an end weight of 2,472 g and a feed conversion ratio of 1.67 for Cobb-type birds.