Multi-resolution spatial join
Speaker: Sun Sai, ITEE
When: 2003-09-11 14:00:00
Venue: 78-420
Host: A/Prof Xiaofang Zhou
Abstract:Multi-resolution is an important topic in the research of spatial
database systems currently. Spatial join is one of the most
important and time-consuming spatial operations. Providing efficient
spatial join algorithm is one of the three necessary conditions of
spatial database system. However, there is still no satisfactory
solution for spatial join processing for multi-resolution. In
multi-resolution spatial database, the result of spatial join is not
required to be 100% accurate. Our goal is to process spatial joins
according to the user requirement of accuracy so that only essential
data will be accessed and computed under this given accuracy
level. To achieve this goal, some pre-processing is performed before
spatial join to ensure the least data retrieval; approximations are
used to represent exact spatial objects to avoid the time-consuming
refinement step. These approximations are much more precise than
those approximations used in traditional spatial join and the
threshold value of approximation quality is computed on the fly
according to users' requirement. Through controlling decomposition
level of approximation, we can control approximation quality and
hence control the error level of filter result to satisfy the user
requirement. An additional benefit is that multi-step filter
strategy may speed up the spatial join between approximations.
Biography:(biography unavailable)
Type: Ph.D confirmation
Contact:A/Prof Xiaofang Zhou, seminar host (zxf@itee.uq.edu.au)
or Guido Governatori (ITEE seminar co-ordinator)
(guido@itee.uq.edu.au)