This dataset includes UAV orthoimages, reference data, and crowd-sourced training data, associated with the paper "From simple labels to semantic image segmentation: Leveraging citizen science plant photographs for tree species mapping in drone imagery." The UAV orthoimages were collected through drone flights and subsequently processed into orthomosaics. For the initial training of CNN models, crowd-sourced plant photographs and species labels were acquired from the iNaturalist and Pl@ntNet platforms. The output from these trained models, alongside the UAV orthoimages, facilitated the generation of training data for the subsequent CNN segmentation models. Please cite this paper if you use this dataset.