Automatic alignment of geographic features in contemporary vector data and historical maps

Duan, Weiwei; Chiang, Yao‐Yi; Knoblock, Craig A.; Jain, Vinil; Feldman, Dan; Uhl, Johannes; Leyk, Stefan

doi:10.1145/3149808.3149816

Cited by 26 publications

(23 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…OldMapsOnline 11 provides a wide variety of georeferenced historical maps covering many places around the globe. These maps can be used to generate training data for geographic feature detection models for historical maps [4].…”

Section: Frameworkmentioning

confidence: 99%

“…In contrast, Kartta Labs defines and integrates independent map processing modules, each with both crowdsourcing and artificial intelligence components, to fully process heterogeneous historical maps. Kartta Labs' pipeline receives a raster map 4 and produces a semantically labeled vector map.…”

Section: Introductionmentioning

confidence: 99%

“…http://github.com/kartta-labs 2 http://openstreetmap.org/ 3 http://mapwarper.net/4 A map in a raster data format, which can be a scanned image, a screenshot of a map, or other types of map images…”

mentioning

confidence: 99%

See 2 more Smart Citations

Kartta Labs

Tavakkol

Chiang

Waters

et al. 2019

Proceedings of the 3rd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery

Self Cite

View full text Add to dashboard Cite

This paper introduces the Kartta Labs project, an ongoing opensource and open-data project aiming at organizing the world's historical maps and making them universally accessible and useful. Kartta Labs' framework is designed as a composition of multiple modules. Each module has a crowdsourcing implementation and an artificial intelligence based implementation. The framework takes images of historical maps, registers them in space and time, generates a vector version of the map content, and allows the users to query for the vector content and recreate the historical maps in various cartographic styles. We refer to this process as unrendering. The resulting machine readable map data can support a variety of scientific studies and applications that require long-term, detailed geographic information in the past, while opening up opportunities in other areas such as entertainment. This paper also presents the preliminary results from one automated module to geolocalize a given historical map.

show abstract

Section: Frameworkmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Kartta Labs

Tavakkol

Chiang

Waters

et al. 2019

Proceedings of the 3rd ACM SIGSPATIAL International Workshop on AI for Geographic Knowledge Discovery

Self Cite

View full text Add to dashboard Cite

show abstract

“…We can further distinguish between orthoimagery [3,4,18] and map images as the alignment target. In the latter category (to which this work also belongs), Duan et al [6] recently proposed a method that adjusts an initial vector-image alignment by searching for consistent local control point deformations that improve the putative alignment. They include a raster foreground extraction method that identifies colors of interest based on a narrow band around the initial vector alignment, resulting in a binarized image for matching.…”

Section: Related Workmentioning

confidence: 99%

Details of Deformable Part Models for Automatically Georeferencing Historical Map Images

Howe¹,

Weinman²,

Gouwar³

et al. 2019

Computer Science

View full text Add to dashboard Cite

Libraries are digitizing their collections of maps from all eras, generating increasingly large online collections of historical cartographic resources. Aligning such maps to a modern geographic coordinate system greatly increases their utility. This work presents a method for such automatic georeferencing, matching raster image content to GIS vector coordinate data. Given an approximate initial alignment that has already been projected from a spherical geographic coordinate system to a Cartesian map coordinate system, a probabilistic shape-matching scheme determines an optimized match between the GIS contours and ink in the binarized map image. Using an evaluation set of 20 historical maps from states and regions of the U.S., the method reduces average alignment RMSE by 12%.

show abstract

“…Key in both cases is the need for abundant and representative training data which requires automated sampling techniques. First attempts in this direction have used ancillary geospatial data for the collection of large amounts of training data in historical maps [34][35][36][37].…”

Section: Recent Developments In Map-based Information Extractionmentioning

confidence: 99%

Map Archive Mining: Visual-analytical Approaches to Explore Large Historical Map Collections

Uhl

Leyk

Chiang

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

Historical maps constitute unique sources of retrospective geographic information. Recently, several map archives containing map series covering large spatial and temporal extents have been systematically scanned and made available to the public. The geographic information contained in such data archives allows extending geospatial analysis retrospectively beyond the era of digital cartography. However, given the large data volumes of such archives and the low graphical quality of older map sheets, the processes to extract geographic information need to be automated to the highest degree possible. In order to understand the salient characteristics, data quality variation, and potential challenges in large-scale information extraction tasks, preparatory analytical steps are required to efficiently assess spatio-temporal coverage, approximate map content, and spatial accuracy of such georeferenced map archives across different cartographic scales. Such preparatory steps are often neglected or ignored in the map processing literature but represent highly critical phases that lay the foundation for any subsequent computational analysis and recognition. In this contribution we demonstrate how such preparatory analyses can be conducted using classical analytical and cartographic techniques as well as visual-analytical data mining tools originating from machine learning and data science, exemplified for the United States Geological Survey topographic map and Sanborn fire insurance map archives.

show abstract

Automatic alignment of geographic features in contemporary vector data and historical maps

Cited by 26 publications

References 15 publications

Kartta Labs

Kartta Labs

Details of Deformable Part Models for Automatically Georeferencing Historical Map Images

Map Archive Mining: Visual-analytical Approaches to Explore Large Historical Map Collections

Contact Info

Product

Resources

About