| 2004 
  
. Web 
ch can 
server 
ORB 
't only 
d. 
r tasks 
h load 
server. 
ta, the 
e. Also 
:W, this 
uilding 
nalysis, 
nology. 
ds. The 
red. If a 
1 active 
| active 
ise, the 
and the 
| thread 
is also a 
est from 
dscape at 
ng model, 
{ which is 
n. Tt well 
the speed 
a structure 
sure 3): 
composed 
Idings. the 
International Archives of the Photog 
inner body, the outer body and the virtual box. The outer shape 
of building will be composed of some LOD shapes. Because 
almost all the surfaces of outer shape of buildings are 
rectangles, these LOS shapes can be constructed by indexed 
quadrangles (index, xi, vi, x», Y2:- Xy. Yn, Zi. Z2), where z, is 
the height of bottom of surface and z, is the height of top 
surface. There are also diverse resolution textures which 
rammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B4. Istanbul 2004 
relative to diverse LOD shapes. According to the distance from 
viewpoint to the position of building, the outer shape of 
building can be used diverse LOD shape. 
  
Building 
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
ID Outer Body Inner Body Bottom Virtual Box Roof 
LOD LOD VRML DEM (GRID) Bounding DEM (TIN) 
Shapel Shape2 3DS Surface Box Surface 
y y 
IndexQuadArray LOD Texture IndexQuadArray IndexTrangleArray 
  
  
  
  
  
  
Figure 3: Data Structure of LOD Building Model 
Because of the complexity shape of building roof, the 
roof can be depicted with DEM (TIN) surface, which is 
composed of indexed triangles array. The bottom of 
building must be suited the city terrain. Generally, the 
city terrain is constructed by DEM (GRID model). The 
inner shape of building will be reconstruct with VRML 
or 3DS file, this file can be linked in web page. The 
building has a virtual bounding box, which is 
approximate shape of building. While the building is so 
far from the viewer, the building will be depicted as his 
virtual bounding box. 
Compared to other 3D city model, LOD building model 
has several advantages for reconstructing web-based 3D 
city landscape. 
a) The vector data to be used to construct the 
building will be indexed triangles array or indexed 
quadrangles array. This will use fewer vector data to 
construct buildings at the same amount of buildings. 
b) The outer shape of buildings has some LOD of 
shapes. It will improve the speed of visualization of lager 
3D city scene. 
¢) The outer shape has multiple images, which are 
used as building texture. The multiple images have diverse 
resolution. 
4. THE NETWORK LOD METHOD 
The network LOD method includes: (1) the multi-resolution 
LOD texture. (2) visibility-based 
dynamic city 
Visualization. 
scene 
1135 
4.1 Multi-resolution LOD texture 
Image data is used to map texture on the surface of 3D building 
model, which is very important in 3D city model. It can 
enhance the reality of 3D city scene. According the user's 
observation habit, the building texture mapping is the farther 
part of 3D building model to the viewer needs lower resolution 
image on its surface, the nearer part of 3D building model to 
the viewer needs higher resolution image. See the figure 4, 
Therefore, the multi-resolution LOD texture data structure is 
proposed. The table I shows the structure of multi-resolution 
LOD texture data. 
  
  
  
421418 
11111 
ipsom 
11111 
  
  
  
  
  
  
  
  
  
  
  
Figure 4. Multi- resolution LOD texture 
In this multi-resolution LOD texture structure, the every 
surface texture of building is created some resolution layer. 
Every layer is created by the different resolution image data 
and compressed by JPEG format. By this way, the texture data 
volume will be reduce and transmitted faster via network.