Automatic vectorization of historical maps: A benchmark
Open Access
- 15 February 2024
- journal article
- research article
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 19 (2) , e0298217
Abstract
Shape vectorization is a key stage of the digitization of large-scale historical maps, especially city maps that exhibit complex and valuable details. Having access to digitized buildings, building blocks, street networks and other geographic content opens numerous new approaches for historical studies such as change tracking, morphological analysis and density estimations. In the context of the digitization of Paris atlases created in the 19th and early 20th centuries, we have designed a supervised pipeline that reliably extract closed shapes from historical maps. This pipeline is based on a supervised edge filtering stage using deep filters, and a closed shape extraction stage using a watershed transform. It relies on probable multiple suboptimal methodological choices that hamper the vectorization performances in terms of accuracy and completeness. Objectively investigating which solutions are the most adequate among the numerous possibilities is comprehensively addressed in this paper. The following contributions are subsequently introduced: (i) we propose an improved training protocol for map digitization; (ii) we introduce a joint optimization of the edge detection and shape extraction stages; (iii) we compare the performance of state-of-the-art deep edge filters with topology-preserving loss functions, including vision transformers; (iv) we evaluate the end-to-end deep learnable watershed against Meyer watershed. We subsequently design the critical path for a fully automatic extraction of key elements of historical maps. All the data, code, benchmark results are freely available athttps://github.com/soduco/Benchmark_historical_map_vectorization.Keywords
Funding Information
- Agence Nationale de la Recherche (ANR-18-CE38-0013)
This publication has 51 references indexed in Scilit:
- Three Things Everyone Should Know About Vision Transformers2022
- ICDAR 2021 Competition on Historical Map Segmentation2021
- Vectorization of Historical Maps Using Deep Edge Filtering and Closed Shape Extraction2021
- Revisiting the Coco Panoptic Metric to Enable Visual and Qualitative Analysis of Historical Map Instance Segmentation2021
- Combining Deep Learning and Mathematical Morphology for Historical Map Segmentation2021
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- Automatic Extraction of Forests from Historical Maps Based on Unsupervised Classification in the CIELab Color Space2013
- Simultaneous Segmentation-Recognition-Vectorization of Meaningful Geographical Objects in Geo-ImagesLecture Notes in Computer Science, 2003
- Off the Shelf Methods for Robust Portuguese Cadastral Map Analysis2003
- Segmentation of thematic maps using colour and spatial attributesLecture Notes in Computer Science, 1998