Automated production of spatial datasets for land categories from historical maps.:Method development and results for a pilot study of Danish late-1800s topographical maps.

This report records the methods and the results of a pilot project aimed at automated production of machine-readable spatial datasets for land categories from Danish topographical maps from the late 1800s. The study was undertaken for two study areas in Jutland, covering around 300 km². Target land categories were: heath, sand dune, wetland, forest and water bodies. The automated geo-data production comprised a combination of object based image analysis, vector GIS, colour segmentation and machine learning processes. Results of an accuracy assessment indicate accuracies that for most categories are around 90 % or higher. A change assessment for the period from the late 1800s until today, revealed a dynamic characterised by decrease in open habitat types due to cultivation and afforestation. We conclude, that automated production of LULC category digital geo-data from historical maps offers a less time consuming and consequently more resource efficient alternative to traditional manual vectorisationThis report records the methods and the results of a pilot project aimed at automated production of machine-readable spatial datasets for land categories from Danish topographical maps from the late 1800s. The study was undertaken for two study areas in Jutland, covering around 300 km². Target land categories were: heath, sand dune, wetland, forest and water bodies. The automated geo-data production comprised a combination of object based image analysis, vector GIS, colour segmentation and machine learning processes. Results of an accuracy assessment indicate accuracies that for most categories are around 90 % or higher. A change assessment for the period from the late 1800s until today, revealed a dynamic characterised by decrease in open habitat types due to cultivation and afforestation. We conclude, that automated production of LULC category digital geo-data from historical maps offers a less time consuming and consequently more resource efficient alternative to traditional manual vectorisatio