DeepTrees 🌳

Deep-Learning based spatiotemporal tree inventorying and monitoring from public orthoimages.

DeepTrees is a Dynamic Platform Project of the Integration Platform 1: “Sustainable future land use” (IP1) at the Helmholz Centre for Environmental Research (UFZ) in Leipzig, Germany.



The project aims to develop and implement deep learning models specifically for tree crown segmentation, tree trait detetion and tree species classification for the Digital Orthomimages Program (DOP) at 20 cm scale in Germany. The resulting tree inventories and monitoring data can be used for various applications, including forest management, biodiversity monitoring, and ecological research.

    📈 Workflows for training and prediction of multiple Deep Learning models.

    📊 Anaylsis Workflows for deriving tree metrics from predictions.

    💿 Training and prediciton datasets of tree crowns.


Supported by:

Helmholtz AI



Cite us:

@software{khan2024deeptrees,
  author = {Khan, T. and Mallast, U. and Knight, T.},
  title = {DeepTrees: Deep-Learning based spatiotemporal tree inventorying and monitoring from public orthoimages},
  url = {https://git.ufz.de/deeptrees},
  year = {2024},
}