The Technology group works on computer vision and robotics, focusing mainly on automating the processes. The processing speed and accuracy of the tasks in those processes is increasing. Nearly every link in the chain, from growing the crop to processing in the food industry, benefits from automating these processes.
A typical task for the robot during the grape harvest is to choose a cluster of grapes from a crate, pick it up and put it into the packaging. In order to do this, the robot has to 'see' the stem of the individual grape clusters, pick up the cluster at the thickest end of the stem, and carefully put it into the packaging. It is possible to accurately detect the clusters using RGB images. Because the colour of the stem is almost identical to that of the grape, the shape (curvature) on the image is also used, in addition to the colour information.
With this automatic stem detection, segments of the stem are shown in an image. Software can then define the cluster stem out of the separate segments, so that the robot knows where the stem is and is able to pick up the cluster. This software is also for the automatic segmentation of stem, leaves and grape, needed for automatic quality assessment of the grapes.
This project is partly sponsored by the Ministry of Economic Affairs, and is part of Top Sector Sustainable Technology and Top Sector Agrifood.
Source: WUR