The next step is to build the face model of the 3-Databases object, i.e. to determine how many faces the 3-Databases object has\, which points define an exact face and the spatial relations of the faces. This is implemented through a consistent labeling algorithm by probabilistic relaxation operations, in which two procedures are involved, the initial probability determination and the relaxation processing. The result of consistent labeling is the face definition for every face. Then, least squares adjustment is performed for all faces simultaneously, fitting the individual faces in an optimal way to the measured points and considering the fact that individual points are usually members of more that one face. This adjustment is amended by observation equations that model orthogonal constraints of pairs of straight lines between boundary points. The details of these algorithms are presented in Gruen, Wang, 1998. Finally, a vector description of 3-Databases objects is obtained, which is represented in a self-developed data structure ( V3D).






CC- Modeler has been successfully implemented on workstations ( Sun SPARC) under X Windows and OSF/ Motif, and has been tested in several projects. The statistics of those data sets are presented in Table 1. ' Structured automatically " refers to the number of roof units that CC-Modeler builds successfully with full automatic processing, and " Structured interactively" refers to the number of roof units that needed to the manually modified in some faces. Obviously, the success rate of CC-Modeler's automatic processing is better that 95% and almost al roof units can be constructed by using the convenient editing tools. The main reason of CC-Modeler failing to process an object is that the measured point cloud was incorrectly coded. The main reasons for editing are measurement errors and ambiguities in topological relations.

For the visualization and animation of the data sets we use various software : AutoCAD Micro station, Inventor, and Polytrim. Figure 2 Shows a view of the city model' Zurich Center " created with Micro Station, including building, rivers, trees and DTM. For photo realistic rendering were combine the vector data of the building and the DTM with images raster data. The raster images are taken from aerial images. Figure 3 shows the city model of " Tokyo Downtown " with the result of mapping image data onto the DTM and some walls and roofs.

Data Structure
V3D is a hybrid data structure. It not only models 3-Databases objects, but also combines raster images and attribute information for each object. The terrain objects are grouped into four different geomatic object types : Point Objects , Line Objects, Surface Objects and Body Objects in V3D, each special object is identified by Type identifier Code ( TIC), referred to as PIC, LIC ,. SIC and BIC, respectively. Tow data sets are attached to each object type: thematic data and geometric data. The images data can be attached to the surface object, body object and DTM object . in fact, the thematic data attributes are built up in a separate data table. It is linked to the object type with a related class label. The definition of the thematic data is user-dependent.

The geometric data set contains the geometric information of 3-Databases objects, i.e. the information of position, shape, size, structure definition ,and image index. The diagram in Fig. 4 shows the logical data structure .


Figure 4: the logical data structures of V3D

For the four object types, four geometrical elements are designed . i.e. Point, Edge, facet and Entity, Point is the basic geometric elements in the diagram. The Point can present a point object. It also can be the begin or end point of an Edge. The edge is a line segment which is an ordered connection between two points: begin point and end point. Further, it can be a straight part of a line object or lie on a facet. The facet is the intermediate geometrical element it is completely described by the ordered edges that define the border of the facet. One or more facts can be related to a surface object or Entity geometrical element. Moreover , facet is related to an image patch. Entity is the highest level geometrical element, and it can carry shape information. An entity is completely defined by its bordering facts. Images data and thematic data are two special data sets. Which are built up in tow separated data tables. Each facet is always related to an image patch through a corresponding link.

Once the attribute table is attached and the TIC is labeled, a geometrical element becomes an object type . the DTM is treated as a special data type. Which is described by a series of facets.

Obviously, the topological relationships between geometrical elements are implicitly defined by the data structure. A point object is presented by a distinct Point element. The line object is described by ordered Edges. The surface object is described by the facet with the informations of image Patches. Similarly, the body object is described by Entity that are defined by the facets. Thus the topological relationships between Point and Edges. Edges and Facet, Facet and Entity are registered by the links between the geometrical elements.

Implementation in A Relational Database
In a relational database the most common object to be manipulated is the relation table. Other objects such as index, views, sequence, synonyms and data dictionary are usually used for query and data access. " Table " is the basic storage structure, which is a two -dimensional matrix consisting of columns and rows of data elements. Each row in a table contains the information needed to describe one instance of the entity , each column represents an attribute of the entity. The data model shown in figure 4 is a logical model, which can be implemented by relational data base technology. Figure 5 shows the relational model of the V3D data structure.



Each object type is defined as a table , shown as the upper row. A point type table includes three terms. The point Identification Code ( PIC) is an identification code for a point type object. The attribute identification (AID) is coded the relate an attribute table. Different types of objects may have different attribute tables. For example the te attribute tables of " tree" may have different thematic definitions than " pole ". The Name of Point (NP) is the identification of a geometric point. Which is used to relate it with a distinct element in a point geometric element. The point table is the most basic geometrical element table , which defines the coordinate position of the geometrical points.

The line type table has similar content as the point type table. The difference is that a line type object is identified by the Line identification Code ( LIC) which is not directly linked with the geometric element table Edge, but linked with an intermediate relational table LIC-NIL and then indexed to the Edge table. The Edge table defines the geometrical element edge, in which each edge ( NIL) is described by the beginning point ( BP) and the end point (EP). The surface type table and body type table have similar terms as the line type table. For each type of object a distinct identification code ( SIC or BIC) is labeled. . both ISC and BIC are linked with a merging geometrical element table, Facet and Entity, Facet and Image are defined. Facet-Entity-Image table has tow links; one is related to the Image table; the other is related to the NIL-SID table. Image table is a basic table, which describes all attributes of images, such as athe image name, format, pixel, Camera parameters, orientation parameters etc. the NIL-SID table is another intermediate table, which defines the corresponding relationships between Facet and Edge. Its NIL column is related to the Edge table. The DTM is treated as a special class, which is related to the NIL-SID table through an intermediate table DID-SID-image.

Based on the relational structure shown on the diagram in Figure 5, the query of a geometrical description of a distinct object type is easily realized. For example, the query " Select the geometrical description of an object with the identification code BIC =202", will first index all Facet identification in the Facet-Entity-Image table by its BIC, and then get all edge name identification ( NIL ) , Finally index the position information of structure points with the help of Edge table and Point table.

The queries of topological relationships are divided into tow types; relationships between the geometrical elements of an object and those between objects themselves. The relationships between the geometrical elements are implicitly defined in the above data structure. Though the internal topology is not directly supplied, users can flexibly deduce the relationships, such as joi8jnt, adjacency, left or right, etc. the queries of topological relationships between objects are not considered in the above data structure because they are application-dependent.



Prototype System
Based on the above data model and structure, a spatial information system, CC-SIS, has been successfully developed and implemented on a workstation ( Sun SPARC ) under X-window, and ORACLE datable . the seven function units of CC-SIS are shown in the Figure 6. it can directly handle the data file generated by CC- Modeler and DXF. The edit function is used for graphic editing, which is to be developed in the future. The view port, zooming, etc. further, three types of rendering are also available , wireframe, shading and texture mapping. The image function supplies the tools for interior orientation of the images in order to map natural texture form image. An example of this function is shown in figure. 7.



The manipulation of data is supplied by the Data function. It includes tow sub-modules; one is used for the operation on layers ; the other is to input the attributes for the selected object . the Geo-query function includes tow tools: geometry query and topology query and topology query. The former is used to query the separated object by the point, line or entity selection ; the latter is employed to query topological relationships between different objects . figure 8 shows the geometric query of CC-SIS . the user can mark an object ( e.g building ) with a cursor . thus triggers and displays the corresponding attributes an geometrical/topological information. The operations on a database are defined in the Database function, including database link and SQL- query. SQL-query is a sub-menu, in which standard SQL queries are supplied.



Conclusion
CC-Modeler is a powerful data acquisition tool for the generation of 3-D city models. Our experiments shows, that it is flexble, robust and accurate . in several projects we have achieved a success rate of better that 95% percent in fully automated structuring. Remaining problems are indicated and can be solved interactively. We have developed our own data structure V3D wit interfaces to a variety of CAD and visualization packages. CC-Modeler cannot only reconstruct multiple kinds of 3-D objects such as buildings, waterways, roads, trees, DTM, etc. but also map images onto these objects. This can be combined with data from general land use, communication systems, utilities, property and administrative boundaries, etc. to generate a complete 3-D city model.

Given an efficient method for 3-D data acquisition, the generation of a powerful 3-D spatial information system becomes even more mandatory. Our prototype system CC-SIS ( CyberCity Spatial information system ) has proven to represent a suitable concept which is worth developing further.

Based on our proprietary V3D vector data structure, a relational data base has been created. The data to be operated on can be logically separated into vector data, image and thematic information. I this paper the focus is on the geometrical part of the database ( vector data and images. ) our pilot application show that V3D is a suitable structure for the representation of 3-D objects, images and thematic data. it is possible the answer most of the questions about topology, position and shape of objects by means of geometric or SQL queries.

References

• Gruen, A., Wang, X., 1998, " CC-Modeler: a topology generator for 3-D city models", int. Archives of Photogrammetry and Remote sensing, Vol. 32, Part 4 , pp 188-196.
• Gruen, A., Baltsavias, E., Hericsson, O., ( eds), 1997. Automated extraction of a man-made objects form aerial and space images (II). Procedings of athe Monte Vertia Workshop, May 1997, Brikhause Verlag, Basel.