Development of a Semantic Segmentation Approach to Old-Map Comparison

This paper describes an innovative computational approach for comparing old maps. Maps olderthan 20 years remain a vast treasure of geographic information in many parts of the world withpotential applications in many environmental and social analyses, e.g., establishing road constructionover the past 80 years or identifying settlement growth since the middle ages. Semantic segmentationhas developed into a viable computational method for analysing old maps from previous centuries.It allows for the discrete identification of elements, e.g., lakes, forests, and roads, from cartographicsources and their computational modelling. Semantic segmentation uses convolutional neuralnetworks to extract elements. With this technique, we create a computational approach to compareold maps systematically and efficiently.