Fraunhofer Institute for Industrial Mathematics ITWMDescription
The thesis develops AI-supported image processing methods to identify conifer (softwood) species from high resolution microscopy images, supporting species protection and enforcement of the Washington Convention (CITES). It evaluates modern AI approaches—vision transformers, foundation models, and classical image processing—to achieve robust, scalable recognition across variable sample quality and preparation. The work is conducted in collaboration with the Thünen Institute of Wood Research within the WoodFiberID project.