GrowliFlower: An image time?series dataset for GROWth analysis of cauLIFLOWER

This article presents GrowliFlower, a georeferenced, image-based UAV time series dataset of two monitored cauliflower fields size 0.39 and 0.60 ha acquired in 2020 2021. The contains RGB multispectral orthophotos from which about 14,000 individual plant coordinates are derived provided. enable the users extraction complete incomplete image patches showing plants. collected phenotypic traits 740 plants, including developmental stage as well size. As harvestable product is completely covered by leaves, IDs provided to extract pairs plants pre post defoliation, facilitate estimations head Moreover, pixel-accurate leaf instance segmentations, stem annotations address tasks like classification, detection, segmentation, similar computer vision tasks. aims foster development evaluation machine learning approaches. It specifically focuses on analysis growth derivation automation agriculture. Two baseline results segmentation at level based labeled data presented. entire set publicly available.