Full text: Proceedings; XXI International Congress for Photogrammetry and Remote Sensing (Part B6b)

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B6b. Beijing 2008 
221 
There are many cardinal relationships of the matched objects 
between old version dataset and new version dataset: 1-1, 1-N, 
N-1,N-M. To get these cardinalities we must make bidirectional 
match between old and new version datasets. The match from 
old version dataset to new version dataset bring 1-N cardinal 
relationship, and vice versa. Multiple object match can avoid 
the inconsistency efficiently during the updating, which can 
describe the holistic change-only information. According to the 
cardinal relationships, the corresponding change events can be 
described as following. 
• 0-to-1: the unmatched objects in new version dataset 
is described by the new created objects; 
• l-to-0: the unmatched objects in old version dataset 
is described by the deleted objects; 
• 1-to-l: one object in old dataset matched by one in 
new dataset is described by 1-1 cardinal relationship; 
• 1-to-N: one object in old dataset matched by many 
objects in new dataset is described by 1-N cardinal 
relationship, which means that the object is split into 
several objects; 
• N-to-1: many objects in old dataset matched by one 
object in new dataset is described by N-l cardinal 
relationship, which means that the objects merge 
into one object; 
• M-to-N: M objects in old dataset matched by M 
objects in new dataset is described by M-N cardinal 
relationship, which means that the objects change 
their structure and regroup; 
• Other objects are stable objects. 
5. DEFINITION OF CHANGE-ONLY INFORMATION 
AND UPDATING 
The change-only information includes not only the changes 
about geographic information, but also the relationships 
between master database and client database and operators in 
database to be implemented in updating. So we represent the 
change-only information in three aspects. The first aspect is the 
difference between spatial objects in semantics, graphics and 
attributes in the details. The changes in fields domain are 
recorded, such as the reshape of road with different coordinates. 
The second component is the relationships among spatial 
objects before and after updating, and we analyze the 
relationships at object level and represent them with change 
events. Also we can define some concepts to describe the 
change of reality features, such land mergence, road split. In the 
third component we express the operators information and rules 
in change-only information at database level in order to carry 
out the implementation of database updates. 
Through the change detection we can get the information about 
geometric, semantic and attribute change about spatial objects. 
In order to satisfy the users’ query about change, the change- 
only information must include the event description to express 
the spatial entities’ changes. Also change-only information will 
be used to update client database ultimately, so the operations 
in database must be considered in change-only information. So 
we definite the change-only information with five components, 
identifier, difference of graphics, semantics and attribute, events. 
[J AAttribute 
AI={ID, Asemantics, Agraphics, <-> , event} 
Because of the particularity of identifier, we list it in change- 
only information. But actually we don’t need the uniform of 
identifiers in two databases, and it can be deal as one common 
item of object. “Asemantics” represents the difference of 
semantics between two spatial objects, which mainly involve 
classification information. “Agraphics” delivers the difference 
of geometric shape of two objects. Event includes not only the 
change information of reality entities, like reshaping, 
abandoning of road, but also the operators with database to 
update, which are showed in table 1 . Figure 4 illustrates the 
result of change-only information with graphics between old 
version dataset and new version dataset. 
Figure 4. Case study: A)old dataset; B) new dataset; C) deleted road; Djadded road; Ejconnected road; Fjreshaped road. 
When we perform the updating we must publish and hand out different application fields the physical storage files or data 
the change-only information to the customers, and there should formats are different. We know that there are log files to record 
be some methods to exchange the updating information. In the changes of database, and we should create exchange file of
	        
Waiting...

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.