CLASSIFICATION AND REPRESENTATION OF CHANGE IN SPATIAL DATABASE
FOR INCREMENTAL DATA TRANSFER
Huaji ZHU'?, Jun CHENS, Jie JIANG?
| The Institute of Remote Sensing Application of CAS, Datun Road, Beijing, China, 100101
zhuhuaji@sohu.com
2 National Geomatics Center of China, 1 Baishengcun, Zizhuyuan, Beijing, China, 100044
Key words: incremental data transfer, change, Geographical Object, update, Classification, representation
ABSTRACT: Nowadays, more and more spatial databases are used in various fields, so the demands for keeping spatial database
"fresh" are growing rapidly. Generally, the end-users get the basic spatial data from professional spatial data producers. There are
mainly two kinds of way for transfer the changed data. One is to batch transfer; the other is incremental data transfer. With the batch
method, the whole up-to-date database is delivered. This process is time-consuming and might induce significant risks of errors
occurrence and information loss.. Recent years, more and more people begin to study incremental updating of spatial database. This
method of updating makes it possible to transfer change-only information to the end-users, namely to transfer the incremental data. In
order to find out the changed objects in a database, firstly we need to identify uniquely all of the objects. In this paper, we define a
geographical object as a new 4-tuple (semantic descriptor, thematic descriptor, spatial descriptor, and temporal descriptor]. Spatial
descriptor consists of geometric component, position component and topologic component. What change happens to an object? In
order to answer this question, firstly, we have to identity and classify changes of geographical objects. Based on the 4-tuple model of
geographical object, we propose taxonomy of the change of geographical object and describe these changes with data/knowledge
packets.
1. INTRODUCTION
Geographic Information Systems (GIS) are increasingly used in
many domains, such as traffic and water sections. Since,
implementing such systems are complex, they are always based
on basic spatial data which are provided by special institutions.
Users, generally, gain reference geographic data from producers
in order to establish their GIS. For instance, a realty company
purchases from a producer a geographic database representing the
land usage of a given region for its company planning application.
Users regard the data provided by producers as spatial reference
data. Geographic data producers are responsible for producing
and maintaining up-to-date databases and delivers new
information to users. In this way, there are many advantages. But
many new problems appear. How to update and maintain user
database, especially maintain the spatial reference data, is one of
the main challenges. When the databases created by the data
producer are updated, the changed data must be transferred to the
end-user to keep their client-databases current. This includes the
management of geographical changes. A more explicit focus on
change has considered the semantics associated with change,
(Claramunt and Theriault 1995, 1996 ). These studies, however,
have omitted certain kinds of change and so far no systematic
treatment of change has been undertaken and they have no given
formal expression. The objective of our research is the
identification of a model for the classification and description of
changes in topological database.
2. TRANSFER OF UPDATES
There are several solutions available for the transfer of updated
dataset (Spéry,2001).
> Bulk data transfer
^ Incremental data transfer: Data producers only provide
those objects that have been changed since the last transfer.
In this case, the producer has to identify geographical
changes in order to generate relevant data sets to each user
context (Wijngaarden et. Al., 1997).
598
Interne
Typica
dissem
in the
rate of
By bul
signifi
databa:
can’t a
mode.
a data
inform
versior
change
spatio-
change
1994: .
model
classifi
geogra
the GI:
Severa
propos
Egenhc
that gei
Hornst
geogra
denote
people
limited
The ob
the cla
explici
change
geogra|
{semar
and sp
taxonoi
AR
Geogra
attribut
by ther
relatior
corresp
1998),
