TY - GEN

T1 - Reconstructing historical rural addresses with VGI and digitized aerial photography

AU - Perner, Mads Linnet

AU - Svenningsen, Stig

N1 - Conference code: 4

PY - 2019

Y1 - 2019

N2 - This paper describes an attempt to develop a historical GIS of farms with metadata from digitized aerial photography. With the current effort of mass digitization, wide ranges of new data sources are becoming available to historical scholars. However, for these digital sources to be included in new scholarship, a prerequisite is the research infrastructure necessary to process them. Our project has examined how metadata on digitized aerial photography, generated by volunteers in the Royal Danish Library’s crowdsourcing effort, might provide a short cut to historical GIS infrastructure, which would otherwise require significant resources to build. As part of the digitization of thousands of aerial photographs of rural properties, the library had the help of volunteers to geo-locate each photograph. As each photograph often represented a single property, the data points and their metadata are representative of a certain address. This paper outlines the steps we took to develop the raw material into a dataset containing locations of historical rural addresses. Based on a pilot study of a limited area, we discuss the quality and accuracy of the data resulting from our approach. We found that the overall quality of data extracted is acceptable compared with the traditional approach of manually plotting in farm-localities by hand.

AB - This paper describes an attempt to develop a historical GIS of farms with metadata from digitized aerial photography. With the current effort of mass digitization, wide ranges of new data sources are becoming available to historical scholars. However, for these digital sources to be included in new scholarship, a prerequisite is the research infrastructure necessary to process them. Our project has examined how metadata on digitized aerial photography, generated by volunteers in the Royal Danish Library’s crowdsourcing effort, might provide a short cut to historical GIS infrastructure, which would otherwise require significant resources to build. As part of the digitization of thousands of aerial photographs of rural properties, the library had the help of volunteers to geo-locate each photograph. As each photograph often represented a single property, the data points and their metadata are representative of a certain address. This paper outlines the steps we took to develop the raw material into a dataset containing locations of historical rural addresses. Based on a pilot study of a limited area, we discuss the quality and accuracy of the data resulting from our approach. We found that the overall quality of data extracted is acceptable compared with the traditional approach of manually plotting in farm-localities by hand.

UR - https://www.mendeley.com/catalogue/3a08df20-37bd-332e-b01f-0446e4ef0ac9/

UR - https://www.mendeley.com/catalogue/3a08df20-37bd-332e-b01f-0446e4ef0ac9/

M3 - Konferenceartikel

VL - 2364

SP - 358

EP - 364

JO - CEUR Workshop Proceedings

JF - CEUR Workshop Proceedings

SN - 1613-0073

Y2 - 5 March 2019 through 8 March 2019

ER -