Synthetic Map Generation to Provide Unlimited Training Data for Historical Map Text Detection

TL;DR

This paper introduces a style transfer-based method to generate unlimited synthetic historical map images with annotated text, significantly improving text detection models' performance on historical maps.

Contribution

The paper presents a novel approach to automatically generate large-scale annotated historical map images using style transfer, reducing manual annotation efforts and enhancing text detection accuracy.

Findings

Synthetic maps improve text detection accuracy.

State-of-the-art models benefit from synthetic training data.

Method reduces manual annotation effort.

Abstract

Many historical map sheets are publicly available for studies that require long-term historical geographic data. The cartographic design of these maps includes a combination of map symbols and text labels. Automatically reading text labels from map images could greatly speed up the map interpretation and helps generate rich metadata describing the map content. Many text detection algorithms have been proposed to locate text regions in map images automatically, but most of the algorithms are trained on out-ofdomain datasets (e.g., scenic images). Training data determines the quality of machine learning models, and manually annotating text regions in map images is labor-extensive and time-consuming. On the other hand, existing geographic data sources, such as Open- StreetMap (OSM), contain machine-readable map layers, which allow us to separate out the text layer and obtain text label annotations easily. However, the cartographic styles between OSM map tiles and historical maps are significantly different. This paper proposes a method to automatically generate an unlimited amount of annotated historical map images for training text detection models. We use a style transfer model to convert contemporary map images into historical style and place text labels upon them. We show that the state-of-the-art text detection models (e.g., PSENet) can benefit from the synthetic historical maps and achieve significant improvement for historical map text detection.