New software will automatically recognize plant traits

The Herbarium Haussknecht in Jena contains a large amount of plant specimens. Recently, hundreds of these specimens have been digitized and are now available as high-resolution digital images. In the context of the newly funded MAMUDS (Managing Multimedia Data for Science) project, a team of scientists from Jena and Tunisia is developing new software to automatically extract information on plant traits from these digitized images, e.g. leaf size and leaf shape.

Once implemented for the digital images from Herbarium Haussknecht, the software is supposed to be able to extract trait data from digitized plant specimens increasingly available from natural history collections all over the world. Extracted data will contribute to quantify plant functional diversity in the global database of plant traits – TRY – managed at Max Planck Institute for Biogeochemistry in Jena by Dr. Jens Kattge. The TRY database is a largely open access global archive with currently 5.6 million curated trait records of about 100,000 plant species.

Jitendra Gaikwad, leader of the iDiv Biodiversity Informatics Group (BDU), coordinates the MAMUDS project, with scientists from the German Center of Integrative Biodiversity Research (iDiv), the University of Jena and the University of Sfax (Tunisia). The MAMUDS project is funded by the Federal Ministry of Education and Research with about 25.000 Euros for two years and the same amount from the Ministry for Research in Tunisia.