AUTOMATED DERIVATION OF A 1:300 000 TOPOGRAPHIC MAP FROM
SWISS DLM VECTOR 200
M. Bobzien1, I. Petzold2, D. Burghardt2
1 - Axes Systems AG, Alpnach, Switzerland
2 - University of Zurich, GIS Department, Zurich, Switzerland
m.bobzien@axes-systems.com
The National
Mapping Agency of Switzerland (swisstopo) is currently undertaking a complete
redesign its map production process in the project Optina-LK. The aim of the
project is the automatic derivation of Digital Cartographic Models (DCM) of
scales between 1:25 000 and 1:300 000. The sources are two Digital
Landscape Models (DLM) of different resolutions with scales around
1:10 000 and 1:200 000. The automated derivation will be performed
within a subproject called SysDab, in which all automated generalisation operations
as well as the automated part of the incremental updating will be performed.
This paper
presents a feasibility study on the derivation of the DCM300 as the basis for a
map of scale 1:300 000 from the Vector200-DLM. The derivation process
consists of several steps of transformation and generalisation, all performed
automatically in one continuous process. The process includes model
transformation from DLM to DCM, as well as various generalisation operations
such as line simplification, line smoothing, aggregation, merging,
displacement, and reclassification. Both models are stored in a newly created
MRDB enriched with meta information about the applied generalisation
operations. This information is used as a prerequisite for automated incremental
updating.
A workflow
management system orchestrates and controls the process. The result is a
completely automatically derived DCM300. Specific cartographic demands can be
fulfilled with manual editing afterwards.
The
derivation process makes use of the results of two research projects – DRIVE
and SerAX. The former focused on the automated generalisation processes with
MRDB and automatic incremental update, the latter – which is still ongoing –
focuses on the orchestration and control of generalisation processes with help
of a workflow management system. This paper shows the results of the automated
generalisation process. It discusses possibilities and limitations of the
automation, including the requirements on the source data.