Geometrical Stem Detection from Image Data for Precision Agriculture.

High efficiency in precision farming depends on accurate tools to perform weed detection and mappingof crops. This allows for precise removal of harmful weeds with a lower amount of pesticides, as well as increaseof the harvest’s yield by providing the farmer with valuable information. In this paper, we address the problem offully automatic stem detection from image data for this purpose. Our approach runs on mobile agricultural robotstaking RGB images. After processing the images to obtain a vegetation mask, our approach separates each plantinto its individual leaves, and later estimates a precise stem position. This allows an upstream mapping algorithmto add the high-resolution stem positions as a semantic aggregate to the global map of the robot, which can beused for weeding and for analyzing crop statistics. We implemented our approach and thoroughly tested it on threedifferent datasets with vegetation masks and stem position ground truth. The experiments presented in this paperconclude that our module is able to detect leaves and estimate the stem’s position at a rate of 56 Hz on a singleCPU. We furthermore provide the software to the community.