×

You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Title
The 3rd ISPRS Workshop on Dynamic and Multi-Dimensional GIS & the 10th Annual Conference of CPGIS on Geoinformatics
Author
Chen, Jun

ISPRS, Vol.34, Part 2W2, “Dynamic and Multi-Dimensional GIS”, Bangkok, May 23-25, 2001
225
e Internet.
i to model
i on web)
■ displayed
)ject in the
at a query
TEM
itecture of
VebGIS is
urce
tem
i-sever, and
VRML files
r-server a n
running for
ough ADO
ie attributes
!l geo-object
r is used for
information
ser plug-in,
irtual scenes
yiodel (DEM)
sreograph by
VirtuoZoNT digital photogrammetry system of the supresoft inc..
Spatial data of this DEM then transformed into VRML format.
The vector information of geographic objects such as buildings
was extracted based the DEM, and also transformed into the
VRML file.
The orthograph of the dalian port area and the surface images of
other geo-objects act as texture. All these images are stored in a
certain directory, ImageTexture node of VRML mapping them to
relative geo-objects. When VRML browser creating virtual scene,
the texture image will be obtained from the certain directory and
mapped to relative modeling, making the scene look more real.
Object in VRML file with more details looks more real, but on the
other hand, VRML browser would take more time to display it. A
Level-of-Detaii (LOD) grouping is one way to optimize the
rendering performance and the display speed of VRML world.
Create LOD that contain multiple copies of the object, called
levels, modeled to various degrees of complexity. There are two
types of LOD, distance and performance. A distance LOD
specifies the distance at which each level is displayed. In the
VRML file of digital harbor, we create four versions of a
geo-object. The close-up is fully detailed. As moving away from
the object, a distance LOD replaces the fully detailed object with
a simpler model, as moving more far away, the model might be a
simple bounding box with color. For the farthest distance, a
empty level make the object disappear. Add these four versions
to a LOD grouping so the browser displays a different model
from the group depending on distance from current viewpoint.
The range value of every level is set according to the attribute
grade of the object. As a concomitant of LOD grouping, multiple
scales spatial information of geo-objects is formed, based on
which multi-scale display is available.
3.2 INFORMATION QUERY
In the digital harbor system, bidirectional information query
based on spatial feature and based on attribute are available.
3.2.1 SPATIALFEATHER BASED INFORMATION QUERY
In VRML file, geographic objects are designed as child of anchor
node, and the URL field of anchor node link to an ASP
programme that would finish the required information query. In
the virtual scene, when user touch a modeling in which he has
interest, the ID of the correlative geographic object would be
transported to the ASP programme at the apply-server, then a
SQL indict was send through ODBC driver to the SQL server
data-base, desired query started. The attribute information of
object would be extracted from attribute database and
downloaded to the browser and displayed as form. Fig2 is the
illustration of the process.
3.2.2 ATTRIBUTE BASED INFORMATION QUERY
In the digital harbor system, an attribute query page is offered, in
which user can organize the query sentence according to
attribute field of object. Data records meeting with the query
conditions would be downloaded by the ASP programme from
SQL server database to browser and displayed as form. When a
record was selected, the correlative geographic object would be
displayed in the virtual scene as conspicuous sight. Fig3
illustrate the process.
Fig3. Illustration of attribute based information query
External Authoring Interface (EAI) technology and Viewpoint
node of VRML were used to finish such attribute based
information query.
For communication between a VRML world and its external
environment an interface between the two is needed, this
interface is call an External Authoring Interface and it defines the
set of functions on the VRML browser that the external
environment can perform to affect the VRML world. The EAI
allows Java applets embedded on an HTML page to
communicate with a VRML scene on that same page. A Java
applet communicate with a VRML world by first obtaining an