3.2 Basic standards
75
In addition, the EOSE-RM includes the following services. Theoretically every IT
service has an equivalent geographie information service. In fact, only three categories are relevant for geographie information:
GI Service types:
1. Geographie Information Human Interaction Service (GIHS). An example is the
interaction of an operator with the graphie interface while picking a line.
2. Geographie Information Model Management Service (GIMS). An example is the
storage of a geographie feature according to an application schema.
3. Geographie Information Processing Service (GIPS). An example is the output of a
map to a screen.
3.2.2 Rules for application schemas (ISO 19109)
The ISO 19100 standards address the full range of geographie information. A certain application only uses a subset of the available standards combined with many
additional details that are beyond the scope of the abstract standards. An example is
a feature cata10gue for a given application.
The ISO 19100 standards provide two approaches to creating a specific application: profiles and application schemas. See chapter 3, section 1.4.1 "Profiles" for further details.
The kernel of ISO 19109 is the definition of a geographie feature. A feature stands
for anything in the real world. For example, a feature can be a single corner stone of
aland parcel, a whole country, a digital elevation model, or a satellite image. In order to integrate a feature into the models of geographie information in a homogeneous way, the ISO 19109 defines the "General feature model" (GFM).
The GFM defines an abstract feature with attributes and operations. Attributes
contain all the static information such as the quality of the feature or its geometrie
properties (point, curve, surface, or solid). Operations contain information about the
change of a feature according to external influences like a road being closed in winter or a road being displayed only within the sc ale range 1 :5,000 to 1 :25,000. This
change is also called the behaviour of the feature.
As the examples show, features can differ in importance and size. In practice, this
often leads to a hierarchical grouping of features. For example, public buildings and
private buildings are both "buildings". The GFM allows for the construction of a
generalization tree where the feature types (class GF _FeatureType) public and private buildings are specialisations within the feature type building.
Generally features reside independently in the dataset. In the case of a 3dimensional dataset the relations between features can be computed in all spatial dimensions. The over-, under-, or level-situation at intersections is a particularly important case for mapping and network computation. A 3-dimensional dataset implicitly contains the necessary information. A 2-dimensional dataset does not allow for
these computations. In order to supply the missing information, the GFM includes
Précédent

- 83/331

Suivant