New AI for Flora Incognita

Prof. Patrick Mäder, Projektleiter von Flora Incognita an der TU Ilmenau "Flora Incognita," Germany's most popular plant identification app, has received new artificial intelligence - as a result, the number of plant species that can be determined has tripled: Around 16,000 species can now be identified worldwide. In addition, the app, which is now available in 20 languages, now also works in offline mode, and its digital educational offering includes a wealth of new information, such as improved distribution maps of many species. Scientists from the Technische Universität Ilmenau and the Max Planck Institute for Biogeochemistry Jena equipped Flora Incognita with an improved technological basis of self-learning deep neural networks. Prof. Patrick Mäder, head of the Data Intensive Systems and Visualization Group and project leader of Flora Incognita at TU Ilmenau, and the research team from Jena have put a lot of effort into developing innovative machine learning training methods for these networks in recent months. "We applied them right away to the Flora Incognita app, and as a result millions of images of plants worldwide could be processed in our data center at TU Ilmenau. With the right images, the new networks are now able to classify many plant species with an accuracy of almost 100 percent." For the new app version, usability and accessibility have also been improved. For example, plant detections can now be created without receiving a net and can later be automatically determined.