Semantic segmentation of historical topographic maps – Diploma thesis by Daniel Schemala

Mr. Daniel Schemala finished his diploma thesis on Semantic Segmentation of Historical Topographic Maps in March. In it, he developed a program that makes it possible to divide scans of historical maps into settlement and non-settlement areas. Those segmented maps were used as a basis for the construction of a geographic information system (GIS). A GIS is used to capture, analyze and visualize geographic objects and their spatial relationships. They are used for the planning and management of natural resources and infrastructures, in transportation, cartography and other areas.

Methods from the fields of machine learning and computer vision were used for this Diploma thesis. The work was supervised by Mr. Hendrik Herold of the Leibnitz Institute for Ecological Spatial Planning (IÖR) and Mr. Dmitrij Schlesinger of the Computer Vision Lab of the TU Dresden. The Diploma thesis was written in German.

Interested? The awarded Diploma thesis by Daniel Schemala is publicly available on Github. You can also download the thesis on „Semantic Segmentation of Historical Topographic Maps“ here.

Would you like to write you final thesis at ScaDS.AI Dresden/Leipzig? We have listed a number of open topics on our website. Of course, you are also welcome to suggest your own topic. Please, send an e-mail with your application for a final thesis at our center to: study.scads.ai@tu-dresden.de.

Check out more news about ScaDS.AI Dresden/Leipzig at our Blog.