The master’s thesis investigates the use of artificial intelligence for the automated vectorization of historical survey sketches as a contribution to improving the quality of the official cadastral map. For this purpose, an AI-based processing pipeline for the detection of cartographic objects is developed, combining tile-based object detection with subsequent geometric post-processing. The results demonstrate that historical survey sketches can generally be evaluated automatically. Downstream processing steps make a significant contribution to improving the overall quality of the results.