Development of a 3D image acquisition system and image processing algorithms for the characterization of the ejection parameters of fertilizer granules. (3DSPREADER)

The aim of this research, a collaboration between ILVO, Ghent University, Agrosup Dijon and the Université de Bourgogne, is to develop a 3-D method to predict the spread pattern of fertilizer grains. Theoretical and practical fine-tuning of the algorithms for the outlet angle and speed lead to the possibility of an intelligent machine with a sensor-based feedback system. The final goal is more sustainable and more profitable agriculture: precise and correct fertilizer distribution in the field is important for environmental as well as economic reasons.

A similar 2-D system has already been developed. Speed and direction in 2-D can now be determined and the spread pattern calculated. Our recent research focuses on 3-D information. We are now developing, testing and validating a method based on stereo matching and motion estimation to estimate the spread pattern from the actual grain trajectory.

A low cost method to determine the grain trajectories can lead to an online tool placed on the spreader itself. This would optimize the distribution of the fertilizer in the field. The algorithms and the methods developed are the basis for an online sensor placed on the spreader which can actively measure and adjust the spread pattern. This innovation will then be made available to agricultural machine developers.