To advance the state of the art in leaf segmentation and to demonstrate the difficulty of segmenting all leaves in an image of plants, we organized the Leaf Segmentation Challenge (LSC). For the challenge we released a training set (which contains raw images and annotations, see example images on the right). Few days before the paper due date, a testing set was released, containing only raw images but no annotations. We will release the training data set here for further investigations.
Download data (260MB) (link no longer available; please visit our dataset page.)
Using the data for scientific publications is allowed and encouraged, given that citations to the following are provided:
Hanno Scharr, Massimo Minervini, Andreas Fischbach, Sotirios A. Tsaftaris. Annotated Image Datasets of Rosette Plants. Technical Report No. FZJ-2014-03837, Forschungszentrum Jülich, 2014
Massimo Minervini, Mohammed M. Abdelsamea, Sotirios A. Tsaftaris, Image-based plant phenotyping with incremental learning and active contours, Ecological Informatics, Available online 6 August 2013, ISSN 1574-9541, http://dx.doi.org/10.1016/j.ecoinf.2013.07.004.
The technical report describes the data acquisition, plant material, and environmental conditions in detail. The paper by Minervini et al. documents the first experiment of the dataset.