Semantic Segmentation for Sequential Historical Maps by Learning from Only One Map

TL;DR

This paper presents a deep learning approach for semantic segmentation of historical maps using only one map for training, leveraging a weakly-supervised age-tracing strategy to improve accuracy.

Contribution

It introduces a novel weakly-supervised fine-tuning method that exploits temporal similarities between maps to reduce the need for manual annotations.

Findings

Achieved a mean IoU of 77.3% on the Hameln dataset.

Improved segmentation performance by approximately 20% over baseline models.

Reached an overall accuracy of 97% in digitizing historical maps.

Abstract

Historical maps are valuable resources that capture detailed geographical information from the past. However, these maps are typically available in printed formats, which are not conducive to modern computer-based analyses. Digitizing these maps into a machine-readable format enables efficient computational analysis. 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 appearance and land-use patterns 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 \textit{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) achieved 77.3\%, reflecting an improvement of approximately 20\% over baseline methods. Additionally, the fine-tuned model achieved an average overall accuracy of 97\%, highlighting the effectiveness of our approach for digitizing historical maps.