Deep learning enables urban change profiling through alignment of historical maps

TL;DR

This work addresses the challenge of extracting fine-grained urban change from historical maps by introducing a modular deep-learning framework that performs dense map alignment, multi-temporal instance detection, and change profiling. It combines self-supervised dense displacement estimation with cartographic guidance and multi-year feature fusion to produce directly comparable building and block representations across years, demonstrated on Paris between $1868$ and $1937$. The method delivers robust alignment, improved instance segmentation under degraded map quality, and informative change trajectories captured at the building-block level, enabling long-term urban morphologies and network evolution analyses. By enabling label propagation, temporal fusion, and transferability to other map series, the framework offers a scalable, automated tool for historical cartography, urban history, and co-evolution studies.

Abstract

Prior to modern Earth observation technologies, historical maps provide a unique record of long-term urban transformation and offer a lens on the evolving identity of cities. However, extracting consistent and fine-grained change information from historical map series remains challenging due to spatial misalignment, cartographic variation, and degrading document quality, limiting most analyses to small-scale or qualitative approaches. We propose a fully automated, deep learning-based framework for fine-grained urban change analysis from large collections of historical maps, built on a modular design that integrates dense map alignment, multi-temporal object detection, and change profiling. This framework shifts the analysis of historical maps from ad hoc visual comparison toward systematic, quantitative characterization of urban change. Experiments demonstrate the robust performance of the proposed alignment and object detection methods. Applied to Paris between 1868 and 1937, the framework reveals the spatial and temporal heterogeneity in urban transformation, highlighting its relevance for research in the social sciences and humanities. The modular design of our framework further supports adaptation to diverse cartographic contexts and downstream applications.

Figures (15)

• Figure 1: Historical maps violate assumptions underlying standard change-detection and image-alignment methods. (a) Two satellite images Chen2020levircd with a time difference of 5-10 years exhibit stable object locations across years, allowing simple overlays to reveal changes. (b) Historical maps from editions a decade apart display obvious object misalignment: object locations shift across sheets (yellow boxes at mosaic boundaries), making direct comparison unreliable. In addition, cartographers frequently adjust text labels for layout or legibility (yellow arrows), introducing displacements unrelated to real-world change. (c) Video frames geiger2013kitti contain smooth, coherent motions where dense-matching algorithms are typically used to align frames, track objects, and estimate per-pixel motion trajectories. (d) Conversely, historical maps exhibit substantial content changes (yellow boxes). These drastic changes, together with deliberate cartographic layout adjustments, undermine the assumptions of dense-matching algorithms, i.e., stable object correspondences for accurate image alignment.
• Figure 2: Method overview. (a) Workflow for historical map alignment and change analysis. Maps covering the same location are first rectified to a common image plane. For map pairs from different editions (years), we estimate a dense displacement field for pixel-level alignment and extract urban instances for each edition. The displacement field is used to align and match instances across years, enabling fine-grained instance-level change analysis. (b) Dense displacement estimation without ground-truth correspondences. A dense-matching network is trained using self-supervision from synthetic transformations with known displacements, and refined with a triplet cycle-consistency constraint linking real and synthetic views. (c) Instance extraction with temporal knowledge. A shared encoder--decoder integrates information from multiple years via temporal feature fusion to predict instance segmentation for the target year. The patch-wise predictions are stitched to produce a complete instance map. (d) Instance-level changes are aggregated into urban blocks, a fundamental spatial unit in urban morphology and planning, to derive block-level change profiles. These profiles enable planning-relevant interpretation of long-term urban morphological change.
• Figure 3: A qualitative comparison of image alignment across different methods. In (a-c), all methods effectively align images despite large displacements between the originals. Our method shows robustness to text displacements in (d) and (e), as well as to object changes in (f-h), maintaining stability even in extreme cases, as seen in (e).
• Figure 4: Average precision score for different objects categorized by their areas: small (area$\leq$500 m$^2$), medium (500 m$^2$<area<2000 m$^2$), large (area>2000 m$^2$), and non-small (area>500 m$^2$). We compare using single maps (Mast2former as the baseline) and using our proposed multi-map fusion strategy. Our model improves the overall accuracy and reduces the interquartile range and number of outliers.
• Figure 5: Qualitative results of our instance extraction method. Individual instances are shown in distinct colors. Our model performs comparably or slightly better to the baseline Mask2Former in (a) building blocks and sidewalks, (b) intricate building structures, and (c) railway neighborhoods, (d,e) while demonstrating clearly improved performance in detecting faded features such as buildings and gardens.