A device for extracting 3D information of fertilizer trajectories

Centrifugal fertilizer spreaders are the most commonly used type in Europe. Their possible inaccuracy results in extra fertilizer being spread, which contributes strongly to environmental imbalance. This imbalance can only be corrected if the correct amount of fertilizer lands in the right place. Accuracy depends on understanding the spreading process from the vane to the soil. This knowledge enables the development of a feedback system that adjusts the spreader settings to the measured spread pattern. The spatial distribution of fertilizer granules on the soil can be predicted by a ballistic flight model based on the measurement of the velocities, directions, pellet size and angular distribution of the pellets. Several models have been proposed to meet this challenge. In an earlier study we proposed a standard camera and a strobe combined with a cross-correlation image-processing algorithm to determine the motion parameters of the pellets at ejection. This system is highly accurate for predicting spread patterns, but only for flat discs, as it only measures the 2D information of the ballistic flight of the pellets. Information on the third dimension will make this model more applicable in practice, as most centrifugal spreaders have concave discs. We developed a 3D image acquisition system based on stereoscopy to characterize the spreading process. Several tests were performed on simulated grain trajectories. Distances and vertical angles were measured with an error of less than 2%