Semantic segmentation of time-series of historical maps by learning from only one map

Historical maps are valuable resources that capture detailed geographical information from the past. However, these maps are typically available in printed or scanned formats, which is not suitable for automatic analyses. Digitizing these maps into a machine-readable, object based, format enables efficient computational analyses. In this paper, we propose an automated approach to digitization using deep-learning-based semantic segmentation, which assigns a semantic label to each pixel in scanned historical maps. A key challenge in this process is the lack of ground-truth annotations required for training deep neural networks, as manual labeling is time-consuming and labor-intensive. To address this issue, we introduce a weakly supervised age-tracing strategy for model fine-tuning. This approach exploits the similarity in land-use patterns and their appearance between historical maps from neighboring time periods to guide the training process. Specifically, model predictions for one map are utilized as pseudo-labels for training on maps from adjacent time periods. Experiments conducted on our newly curated Hameln dataset demonstrate that the proposed age-tracing strategy significantly enhances segmentation performance compared to baseline models. In the best-case scenario, the mean Intersection over Union (mIoU) reached 77.3%, reflecting an improvement of approximately 20% over baseline methods. Additionally, the model achieved an average overall accuracy of 97%, highlighting the effectiveness of our approach for digitizing historical maps.