The design of an integrated system is described for combining spatial and nonspatial attribute data to allow fast query handling. By fast it is meant that search queries can be executed in logarithmic time. Nonspatial attribute data is managed by standard database techniques. Efficient and robust spatial data structures are used to store spatial information. In addition to the standard access methods usually provided for accessing attribute data (i.e. through the data base management system), data can also be accessed through the spatial data structures in a homogeneous way that is transparent to the user. The design makes use of an SQL-like interface. The SQL query language is extended to provide efficient access methods to spatial structures, and to match functional and efficiency requirements desired from a combination of spatial and nonspatial information. The simplicity of SQL helps to clarify the design approach. However, other interfaces can be used with minor modifications.
[1]
Hanan Samet,et al.
Connected Component Labeling Using Quadtrees
,
1981,
JACM.
[2]
Christos Faloutsos,et al.
Fractals for secondary key retrieval
,
1989,
PODS.
[3]
Clifford A. Shaffer,et al.
QUILT: a geographic information system based on quadtrees
,
1990,
Int. J. Geogr. Inf. Sci..
[4]
Christos Faloutsos,et al.
An Efficient Pictorial Database System for PSQL
,
1988,
IEEE Trans. Software Eng..
[5]
Jack A. Orenstein.
Spatial query processing in an object-oriented database system
,
1986,
SIGMOD '86.
[6]
Azriel Rosenfeld,et al.
Digital Picture Processing
,
1976
.
[7]
Thomas C. Waugh,et al.
The GEO VIEW design A relational data base approach to geographical data handling
,
1987,
Int. J. Geogr. Inf. Sci..