Environmental Engineering Reference
base for building a virtual environment. To obtain a suitable 3D dataset of an area of
interest, users of the system can select an area on a 2D digital map in a GIS and the
application sends the matching data to the landscape visualisation component.
The generation of the 3D model and its rendering is done by this software component,
that we have called here SLE. Thanks to this virtual environment, a post-model
analysis can be conducted to address issues specific of the landscape, with an objective
of participative planning and negotiation with the various stakeholders.
The challenge of the SLE software is, without proprietary tools or database
preparation, to extract specific data (elevation, imagery and land cover) from a GIS,
fuse this data in a procedural manner to enhance its apparent quality, and add
vegetation objects to the scene.
Existing Software Tools
Today, 3D landscape visualization software such as Vue® (E-on Software, http://
com/ ) are available. They provide the possibility for modelling and rendering
landscapes with a high degree of realism, but are specifically designed for infographists.
Since 2000, Visual Nature Studio® (3D Nature) is a GIS-compatible version of
World Construction Set® and in 2003; an add-on has been released that introduced
real-time capability on a lower level of detail.
3D plant modelling and landscape design has released LandSIM3D®, a simulation
and design software for quickly modelling existing landscape in 3D from GIS data
sources. It permits to import a project, visualize its integration in a real site, study
its alternatives, its future evolution and decide for the best options.
These software tools are actually used for landscape planning. However, these
are commercial software and not designed especially for research scientists. That is
why we developed a free GIS-compatible and interactive landscape visualisation
software, based on available open-source components.
The representation of 3D landscape models requires a variety of components and
corresponding spatial data types. These include terrain texture (orthoimagery, raster
maps), digital height models (DTM, digital terrain model or DEM, digital elevation
model, a digital representation of ground surface topography or terrain; or DSM,
digital surface model, a topographic model of the reflected surface of the earth
that can be manipulated by computer programs), vector-based 2D geo-objects,
3D objects, object textures, animations and hyperlinks. The spectrum of these
components ranges from very large spatial objects to large numbers of complex and